Every entry in the library about who technology serves and who it harms — the research, the organizations doing the oversight work, the publishers covering the beat, and the writers worth following.
Free search across academic papers, theses and citations, with links to full text where it is public.
From the site: Google Scholar provides a simple way to broadly search for scholarly literature. Search across a wide variety of disciplines and sources: articles, theses, books, abstracts and court opinions.
Why I recommend it: The fastest way to check whether a claim in a news article traces back to an actual paper. Set an alert on a topic and it will email you new work.
An OpenAI-compatible API for unrestricted language models aimed at red teaming, security research, evaluations, and synthetic data, paired with a policy gateway for per-project keys, audit logs, and no data retention.
Why I recommend it: I keep this in the ethics shelf on purpose. Seeing how guardrails get removed for testing is the clearest way to understand why they matter in the tools you actually use at work.
An autonomous AI agent for penetration testing and security research, running through one command-line interface across several major models.
Why I recommend it: If you are moving toward security work, tools like this are what the job looks like now. Learn the agent, but learn the fundamentals it is automating too.
A public demo of Drummer, an experimental 542-million-parameter language model trained from scratch, with chat, continuation and live tool-calling tests.
Why I recommend it: Useful if you want to see plainly what a small, honestly-labeled model can and cannot do.
Transform any topic into peak LinkedIn thought leadership guaranteed to make your followers shudder.
Why I recommend it: I include CringeBot 3000 as a gentle warning: generative AI can make your LinkedIn presence sound impressive and hollow at the same time. Use it to see what over-polished "thought leadership" looks like, then write something that actually sounds like you.
Startup building a way for different AI models to exchange knowledge directly, without translating everything back into text prompts.
Why I recommend it: Early-stage and unproven, but worth watching: model-to-model communication is the kind of shift that quietly changes which technical skills matter.
Global nonprofit defending and extending digital rights for everyone.
From the site: Digital rightsfor everyone. Derechos digitalespara todas. .الحقوق الرقمية للجميع Droits numériquespour tous. Access Now defends and extends the digital
UK-based independent research institute (established by the Nuffield Foundation) ensuring data and AI work for people and society.
In plain terms: This independent research website examines the ethical and legal impacts of artificial intelligence and data. You can read policy reports, explore industry analysis, and attend events to understand how emerging technology affects society.
Why I recommend it: Clear, public-interest research with plain-language summaries.
NYU-based research institute examining the social and political implications of AI, publishing influential annual reports on AI's societal effects.
In plain terms: This research institute analyzes the social, economic, and workplace impacts of artificial intelligence. You can read free reports, policy toolkits, and expert analyses to better understand how AI affects society and the economy.
Why I recommend it: Their reports connect AI directly to jobs and worker power.
Ongoing research archive on how AI and automation affect wages, workers, and economic power in the U.S.
In plain terms: This research archive provides articles and reports on how artificial intelligence affects the workforce. You can explore these studies to learn how new technologies and automation impact jobs, wages, and worker protections.
From the site: Content archives for the Washington Center for Equitable Growth’s work on AI, tech, & the economy.
Why I recommend it: Use this when you need real numbers on AI and jobs for a proposal or interview.
Founded by Joy Buolamwini, AJL combines art and research to expose racial and gender bias in AI and mobilize advocates, researchers, and industry toward more accountable algorithms.
In plain terms: This organization researches and exposes bias and discrimination in artificial intelligence systems, including automated hiring tools. You can explore educational materials, learn about the social impacts of technology, and report unfair automated practices.
Why I recommend it: Start here if you have ever been misjudged by an automated system — including a hiring one.
Non-profit research organisation working on theoretical alignment and on evaluations that test what frontier models are capable of, with public reports.
From the site: ARC is a non-profit research organization whose mission is to align future machine learning systems with human interests.
Why I recommend it: Their evaluations work is why "dangerous capability testing" is now a normal phrase. Read the reports, they are short.
Non-profit building the responsible tech community, with free reports, career guides, mentorship programmes, a job board and a large directory of people working on tech and society issues.
From the site: We
Why I recommend it: One of the best free entry points if you want to move into responsible tech — the reports and community directory are open to anyone, no membership needed.
An AI safety lab focused on "scheming" — models that pursue their own goals while appearing aligned. Publishes research on detecting deception, evaluations and governance advice.
From the site: Apollo Research is focused on reducing risks from scheming frontier AI. Our goal is to secure frontier AI systems across development, deployment, and governance.
Why I recommend it: Their research and blog are free to read. Apollo also sells a monitoring product, so read claims about their own tool as company claims.
Harvard University center studying the ethics, governance, and societal impact of the internet and AI, and anchor institution for the Ethics and Governance of AI Fund.
In plain terms: This research center explores how artificial intelligence and the internet impact society, law, and ethics. You can read free policy publications, watch educational videos, find public events, and check for open job or fellowship opportunities.
Why I recommend it: Decades of open research and fellowships, much of it free to read.
Nonprofit working to align technology design with human wellbeing, addressing extractive incentives in tech and AI.
In plain terms: This nonprofit organization provides educational materials, policy guides, and research on the societal impact of artificial intelligence and social media. You can take courses, listen to podcasts, and use design toolkits to learn about ethical technology practices.
Why I recommend it: Practical framing for anyone rethinking their relationship with their devices.
Cambridge research centre studying risks that could threaten humanity's long-term future, with open papers, seminars and policy submissions.
From the site: We study existential and global catastrophic risks & foster a worldwide community of academics, technologists and policy-makers working to mitigate these risks.
Why I recommend it: Academic and careful. Their reading lists and seminar recordings are the fastest way into the field's actual literature.
The Royal Institute of International Affairs, an independent policy institute based in London, publishing research on international affairs, governance and technology.
From the site: Chatham House, the Royal Institute of International Affairs, is an independent policy institute based in London. Discover what we do, visit our website today.
#governance#international affairs#policy#regulation#research#think tank
Chatham House – International Affairs Think TankAdded Sep 18, 20260 opens
The institute behind the Millennium Prize Problems, with free lecture videos, published proofs, historical mathematics archives and details of its research programmes.
Why I recommend it: Free access to serious mathematics — the lecture library alone is worth bookmarking if you are studying or teaching math.
Independent, community-rooted AI research institute founded by Timnit Gebru, studying the real harms of AI instead of the hype.
In plain terms: This independent institute studies the real-world harms and community impacts of artificial intelligence. You can explore their research publications, learn how technology affects diverse groups, and look for open career opportunities.
From the site: The Distributed AI Research Institute is a globally distributed organization of academics, activists, and engineers conducting community-rooted research.
Why I recommend it: Start here if you want the research-backed counterweight to AI marketing.
Independent nonprofit researching the social implications of data-centric and automated technologies, informing policy and public understanding.
In plain terms: This nonprofit research institute studies how artificial intelligence, automation, and data technologies affect work and society. You can read free reports, guides, and articles or attend public events to understand how emerging technology impacts labor and the economy.
Why I recommend it: Excellent on automated management and surveillance at work.
A Canadian non-profit offering free, structured support for people and families harmed by social media and heavy online life — compulsive screen use, comparison culture, distress after posting or after deleting accounts, financial fallout, and difficulty leaving platforms. Peer-led and recovery-oriented, delivered virtually across Canada. Founded by Kayleigh Robertson.
Why I recommend it: Free, and it says so plainly: all services free, virtual, Canada-wide, currently accepting clients through an intake form. Two honest limits — it is Canada-only, and it is peer-led support grounded in lived experience and emerging research, not clinical treatment, so it sits alongside a doctor or therapist rather than replacing one. Worth knowing if the job hunt itself is what has you stuck online all day.
Company building thermodynamic computing hardware it claims is far more energy efficient than GPUs.
From the site: Building thermodynamic computing hardware that is radically more energy efficient than GPUs.
Why I recommend it: The energy cost of AI is becoming the constraint, and this is one bet on solving it in hardware. The efficiency claims are the company's own and unproven at scale.
Non-profit AI safety research lab publishing technical work on model robustness and evaluation, plus events and a fellowship pipeline for researchers entering the field.
From the site: FAR.AI is an AI safety nonprofit advancing technical research across robustness, deception, and red-teaming to ensure AI systems remain safe and beneficial.
Why I recommend it: Look at their fellowships and events pages, not just the papers — that is where the actual entry points are.
Global foundation committed to social justice, funding technology initiatives that promote equity, inclusion, and accountable AI governance.
In plain terms: This global foundation funds organizations and individuals working on social issues, workers' rights, and technology. You can search for grant opportunities, apply for fellowship programs, and read research reports on the future of work.
Why I recommend it: A major funder of public-interest technology work worth tracking.
Non-profit working on risks from advanced technology, publishing policy work, the AI Safety Index, open letters and a large free podcast and newsletter archive.
From the site: FLI works on reducing extreme risks from transformative technologies. We are best known for developing the Asilomar AI governance principles.
Why I recommend it: Their AI Safety Index is the most readable scorecard of what the big labs actually do about safety. They campaign, so read the policy pages as arguments.
Research institute publishing free reports and analysis on AI governance, compute policy, security and international coordination.
Why I recommend it: Read this alongside the industry labs' own publications — it covers the governance side that vendor blogs tend to skip. All research is free to download.
#ai policy#governance#regulation#research#security#think tank
Institute for AI Policy and StrategyAdded Sep 21, 20260 opens
Alphabet company applying AI models to drug discovery, spun out of Google DeepMind.
From the site: Isomorphic Labs is building a future where frontier AI can help to unlock deeper scientific insights, faster breakthroughs, and life-changing medicines.
Why I recommend it: A concrete example of AI applied to something other than chat. Worth a look if you are science-trained and wondering where AI hiring is happening outside big tech.
The nation's Black think tank, with a technology policy program focused on equity in the digital economy.
In plain terms: This research organization provides reports and data on workforce policy, technology, and economic issues affecting Black Americans. You can explore their research briefs, reports, and events to learn about labor trends and policy solutions.
From the site: About The Joint Center for Political and Economic Studies is a 501(c)(3) non-profit organization based in Washington, D.C. that creates ideas to improve the socioeconomic status and civic engagement of African Americans. Founded in 1970 to support newly-elected Black officials who were moving from civil rights activis…
Why I recommend it: Their tech policy team publishes work you can cite and hires people from nontraditional paths.
Social Science Research Council program funding and publishing work on technology, power, and public life.
In plain terms: This research platform publishes articles and essays exploring how technology and artificial intelligence affect workers and society. You can read expert reviews and analyses to learn about labor protections, tech ethics, and workplace automation.
From the site: The Just Tech Platform is a forum, catalogue, and showcase for researchers and practitioners at the nexus of technological development, inequity, and social justice.
Why I recommend it: Great source of fellowships and calls for proposals if you want funded research work.
A research nonprofit that independently evaluates frontier AI models to measure what they can actually do and what risks that creates. Reports are free.
From the site: METR is a research nonprofit that evaluates frontier AI models to inform the public about their risks and capabilities.
Why I recommend it: One of the few independent evaluators. Read their reports before you trust a lab's own capability claims.
Microsoft's research division: published papers, open datasets and tools, plus its internship and residency programs.
From the site: Explore research at Microsoft, a site featuring the impact of research along with publications, products, downloads, and research careers.
Why I recommend it: The publications are free to read and the programs page lists real entry routes into research work. Both are more useful than the marketing pages.
Montreal research institute founded by Yoshua Bengio: publications, research teams, and programs for students and visiting researchers.
From the site: Mila is a Montreal-based artificial intelligence research institute that brings together researchers from Université de Montréal, McGill University, Polytechnique Montréal and HEC Montréal.
Why I recommend it: Check the students and programs pages if you want to move toward research work. Academic institutes publish their entry routes more openly than companies do.
MIT research group studying how digital technology and AI change work, wages and productivity, with published papers and reports.
From the site: The MIT Initiative on the Digital Economy (IDE) explores how people and businesses will work, interact, and prosper in the digital era.
Why I recommend it: This is measurement rather than prediction, which is rare in this subject. Go here when you want numbers on automation instead of opinions about it.
MIT research lab exploring human-AI collaboration, alongside a joint fund with Berkman Klein supporting research on AI's ethical and governance challenges.
In plain terms: This academic research site shares news and projects focused on emerging technology, design, and artificial intelligence. You can explore articles on AI ethics and human collaboration, view new inventions, and find related job opportunities.
Why I recommend it: Browse their projects when you want to see what humane technology looks like in practice.
The US National Security Agency's AI Security Center: published guidance, joint advisories and best-practice sheets on securing AI systems against attack. Free to read and download.
Why I recommend it: Practical, specific security guidance you can hand to a technical team. It is written from a national-security standpoint, so priorities reflect government threat models more than everyday consumer risk.
Philanthropic investment firm supporting organizations that harness technology to empower individuals and communities responsibly.
In plain terms: This philanthropic investment firm focuses on responsible technology and its impact on society. You can check their careers page to search for open jobs and learn about their work.
Why I recommend it: Follow their funding to see which responsible-tech ideas are gaining ground.
The stewards of the Open Source Definition, with plain-language explanations of every approved licence and guidance on choosing one for your own project.
Why I recommend it: Read this before you publish code or a template — the license you pick decides what others can legally do with your work.
University of Oxford institute researching the ethical problems arising from AI, from societal downstream effects to how AI systems reflect human values.
In plain terms: This academic website shares research and analysis on the ethical and social impacts of artificial intelligence. You can read publications, attend public events, and search for fellowships, scholarships, and job openings.
Why I recommend it: Philosophy-forward work — useful when you need the "why", not just the "how".
Multi-stakeholder nonprofit coalition of tech companies and civil society organizations shaping best practices and public dialogue on AI's benefits and risks.
In plain terms: This nonprofit website shares research and guidelines on artificial intelligence from tech companies and community organizations. You can explore free reports and frameworks to learn how AI affects the economy, workplace practices, and technology safety.
Why I recommend it: Where industry and civil society actually sit at the same table.
Non-profit helping build the field of digital minds, supporting research and education on AI consciousness, moral status and AI minds.
From the site: PRISM is a non-profit helping to build the field of digital minds, supporting research and education on AI consciousness, moral status, and AI minds.
Develops and advocates for policies that reduce the risk of severe harm from advanced AI, promoting transparency, accountability and safe development.
From the site: We develop and advocate for policies that reduce the risk of severe harm from advanced AI. Our work promotes transparency, accountability, and safe development.
Interdisciplinary Stanford research institute studying how digital technology and AI change work, productivity and shared prosperity, with public papers and data.
From the site: The Stanford Digital Economy Lab is an interdisciplinary research institute shaping a future where technology drives human well-being and shared prosperity.
Why I recommend it: Free to read. Useful when you want measured research on AI and jobs instead of headline claims — check the publication date on each paper, the field is moving fast.
Interdisciplinary Stanford institute advancing AI research, education, policy, and practice to improve the human condition, with strong ethics and governance focus.
In plain terms: This university center shares research, policy updates, and educational resources focused on artificial intelligence. You can browse an AI glossary, read industry reports, and explore professional courses or research fellowships.
Why I recommend it: Their policy briefs are readable and free — a good first stop if AI governance feels opaque.
A nonprofit that works to make the tech industry share its prosperity and answer for economic harms. It focuses on housing and working conditions, and publishes research such as the 2026 California AI Compass.
Why I recommend it: An advocacy group, so its reports argue a position. Its research on contract workers and AI in the workforce is useful for anyone weighing a tech job.
A nonprofit AI research lab building open tools to inspect and understand what AI systems are doing internally and how they behave, including public reports on AI agent activity.
Research center developing AI systems that are provably beneficial and aligned with human values.
In plain terms: This university research center focuses on creating safe and beneficial artificial intelligence. You can read published research papers, follow recent news and blog updates, and explore opportunities to work with their team.
Why I recommend it: Technical AI safety, explained by the people who defined the field.
A media platform covering Black professionals in technology, with news, career content, and conference programming.
Why I recommend it: One of the clearest places to see who is building and hiring in tech beyond the usual coverage. The career section is more useful than most tech media.
A long-running technology news publication with careful, technical reporting on computing, science, policy and security — deeper than most tech headlines and clear about what is known versus claimed.
From the site: News and reviews, covering IT, AI, science, space, health, gaming, cybersecurity, tech policy, computers, mobile devices, and operating systems.
Why I recommend it: One of the few tech outlets that reads a filing or a paper before writing about it. Free to read, with an optional paid subscription that removes ads.
The open-access preprint server for physics, mathematics, computer science, and related fields — a primary source for cutting-edge AI and machine-learning research papers.
Why I recommend it: The best place to read AI research before it hits journals or the press; search by tag or author to follow a specific line of work.
Power dynamics and inside stories from Big Tech, by Alex Kantrowitz.
In plain terms: This newsletter and podcast covers inside reporting on major tech companies and their impact on society. You can read weekly updates and listen to interviews to stay informed about the technology industry.
Why I recommend it: Good on how decisions inside big companies land on workers.
Independent cybersecurity and technology news site covering malware, ransomware, data breaches, privacy, and support guides for Windows, Linux and macOS.
From the site: BleepingComputer is a premier destination for cybersecurity news for over 20 years, delivering breaking stories on the latest hacks, malware threats, and how to protect your devices.
Why I recommend it: Free to read and ad-supported. Use it as a practical incident tracker and a source of plain-language security guidance, not as a single source for attribution.
Brookings' running collection of technology and innovation research, including AI policy, labour effects and governance. Free to read.
Why I recommend it: A steady source of policy research when you want something more careful than news coverage. Brookings is a think tank with its own funders and viewpoints.
Free journalism and analysis on nuclear risk, climate change and disruptive technologies including artificial intelligence, from the group behind the Doomsday Clock.
Why I recommend it: Good grounding if you want to argue about AI risk with facts rather than vibes.
Official news from CERN, the European particle physics laboratory: accelerator and experiment updates, computing and engineering write-ups, and knowledge-sharing pieces, filterable by topic and audience (general public, students, educators, policymakers). Free to read, no account.
Why I recommend it: Worth following if you want technology news written by the people doing the work rather than by a tech press cycle. The Computing and Knowledge sharing topics are the useful filters for a career audience — CERN publishes plainly about large-scale computing, data handling and open-source work, and the writing is aimed at non-specialists.
Daily writing and research publication covering AI, business strategy, and how knowledge workers actually use new tools, plus its own suite of AI products.
Why I recommend it: I read Every when I want thinking about AI that goes beyond hype cycles. The essays are long but they change how you work.
Consumer technology news covering AI, gadgets, science, and the culture around them.
Why I recommend it: Good for staying conversational about tech trends without a paywall. Skim headlines weekly so interview small talk about your industry stays current.
Newsletter reporting on the fight to reshape technology in the public interest.
In plain terms: This newsletter reports on efforts to reshape technology in the public interest. You can read regular articles to stay informed about technology ethics and industry reform.
From the site: Idea Trafficking. Click to read Hard Reset, by Trafficker 01, a Substack publication. Launched 5 years ago.
Why I recommend it: Skimmable and current — good for staying briefed in ten minutes a week.
Engineering and technology news site covering AI, energy, space and hardware, free to read with an optional paid ad-free tier.
From the site: Explore Interesting Engineering for cutting-edge articles, news, and insights on technology, innovation, and the future of engineering worldwide.
Why I recommend it: Broad and fast-moving, sometimes breathless. Fine for spotting stories, worth checking the original source before repeating one.
Workplace and tech news site covering AI, careers, startups and hiring, with a focus on the Indian tech industry.
From the site: OfficeChai is India’s top platform dedicated to the workplace, office buzz and Indian business. We profile interesting startups, feature inspiring career stories, break news from the corporate world, and provide a glimpse of the work culture at top organizations. - See more at: https://officechai.com/about/#sthash.deB…
Why I recommend it: Useful for AI hiring and workplace stories that Western outlets skip. Short posts, so treat them as pointers.
Reporting on the intersection of Silicon Valley and democracy, by Casey Newton.
In plain terms: This publication delivers reporting and analysis on artificial intelligence, social platforms, and the tech industry. You can read free articles to stay updated on how new technology impacts modern work and business.
Why I recommend it: Independent accountability reporting on the platforms we all depend on.
Open-access academic journal publishing peer-reviewed research on robotics, automation, and their economic and social consequences.
Why I recommend it: Free peer-reviewed research on automation. Denser than a blog post, but the citations are gold if you are writing or speaking on this.
Analysis of the strategy and business side of technology and media, by Ben Thompson.
In plain terms: This website provides articles and podcasts that analyze the business strategy and impact of technology companies. You can explore in-depth commentary to better understand industry trends and how modern tech businesses operate.
Why I recommend it: Teaches you to read industry news as strategy instead of headlines.
Nonprofit publication covering the intersection of technology, platforms, and democratic institutions.
In plain terms: This nonprofit publication provides news, opinion, and analysis on how technology impacts government and democracy. You can read articles and listen to podcasts to stay informed on tech laws, platform regulations, and artificial intelligence ethics.
From the site: Tech Policy Press is a nonprofit media and community venture intended to provoke new ideas, debate and discussion at the intersection of technology and democracy. We publish opinion and analysis.
Why I recommend it: They publish outside contributors — a real place to build a byline in this field.
Analysis at the intersection of finance and technology trends, by Byrne Hobart.
In plain terms: This newsletter provides in-depth articles analyzing trends, strategies, and news across the technology and finance industries. You can read detailed company profiles to understand market shifts and explore an included job board.
Why I recommend it: Dense, but it explains where the money behind tech is actually going.
A long-running cybersecurity news site covering breaches, vulnerabilities and, increasingly, AI-related security incidents.
From the site: The Hacker News is the top cybersecurity news platform, delivering real-time updates, threat intelligence, data breach reports, expert analysis, and actionable insights for infosec professionals and decision-makers.
Why I recommend it: Free to read with ads and newsletter prompts. Good for spotting AI security stories early; headlines can be urgent in tone, so check the underlying advisory.
A publisher covering cloud native, DevOps, open source, and AI-native software engineering news and analysis for developers, platform engineers, and engineering leaders.
Why I recommend it: One of the few tech news sources I trust to go deeper than the press release. If you are trying to understand what is actually happening in AI-native engineering, start here.
The Verge's technology section: daily reporting on the companies, products and policies shaping the industry.
From the site: The latest tech news about the world’s best (and sometimes worst) hardware, apps, and much more. From top companies like Google and Apple to tiny startups vying for your attention, Verge Tech has the latest in what matters in technology daily.
Why I recommend it: Free to read and readable. Good for keeping current on the companies you might interview with.
Project and publication examining what we teach AI systems and what those choices say about us.
In plain terms: This project shares anonymous handwritten notes about people to examine what humans teach artificial intelligence systems. You can read the publication to reflect on the personal choices and ethics behind modern technology.
From the site: Anonymous handwritten notes about people
Why I recommend it: Useful for language and framing when you explain AI risk to non-technical people.
Long-running technology publication covering AI, security, policy, and the business of tech.
Why I recommend it: The free articles alone are enough to track where AI and security policy are heading. Bookmark one story a week that touches your field and save the takeaway.
Independent technology journalism founded by Jason Koebler, Emanuel Maiberg, Samantha Cole and Joseph Cox. Covers tech, AI, privacy and internet culture.
From the site: 404 Media is an independent media company founded by technology journalists Jason Koebler, Emanuel Maiberg, Samantha Cole, and Joseph Cox.
Why I recommend it: Some articles may be behind a membership paywall; the homepage and many stories are free to read.
Online magazine of ideas and culture publishing long-form essays on philosophy, science, ethics, technology and society. Free to read.
From the site: Aeon is a magazine of ideas and culture. We publish in-depth essays from the world’s most incisive and ambitious thinkers, and a mix of original and curated videos — free to all.
Daniel Kokotajlo's research blog on what a world with very capable AI might look like, including the AI 2027 scenario work.
From the site: Preparing for a world with AGI. Click to read AI Futures Project, by Daniel Kokotajlo, a Substack publication with tens of thousands of subscribers.
Why I recommend it: This is forecasting, not measurement — treat it as a well-argued guess. Useful for the questions it raises rather than the dates it puts on them.
A guided system where AI builders share insights, contribute projects, evaluate real-world impact, and amplify practical skills alongside an AI mentor.
Why I recommend it: I like the emphasis on building and evaluating impact rather than just consuming AI news. Useful if you want to move from "AI curious" to "AI capable."
The UK AI Security Institute blog, sharing research and work to enable advanced AI governance.
From the site: View AISI research and work. The AI Security Institute is a directorate of the Department of Science, Innovation, and Technology that facilitates rigorous research to enable advanced AI governance.
Research commentary from the UK's national institute for data science and AI, covering AI safety, public-sector deployment, health data and the social impact of automated systems.
Why I recommend it: Solid, evidence-based writing on AI policy — a useful counterweight to vendor blogs.
Newsletter and blog covering transformative AI risk, AI alignment, forecasting, longtermism and how to navigate the century ahead. By Holden Karnofsky; posts are freely readable with an optional audio version.
From the site: For audio version, search for "Cold Takes Audio" in your podcast app
A practical blog series from Bian Jiang documenting real workflows for integrating generative AI into daily work, from writing to research to automation.
Why I recommend it: I keep pointing clients to concrete "here is how I actually use it" examples rather than hype. This series is calm, tactical, and honest about what works.
Microsoft's own announcements and positions on AI, work and policy.
Why I recommend it: This is the company speaking for itself, which makes it a primary source and an advertisement at the same time. Useful for knowing what it is committing to publicly.
Brookings fellow writing on AI, workers, and the future of good jobs.
In plain terms: This newsletter features writing from a Brookings fellow about artificial intelligence, workers, and the future of jobs. You can read it to stay informed about how technology impacts the modern workplace.
From the site: Click to read Molly Kinder on Substack. Launched 15 days ago.
Why I recommend it: Her worker-first framing is exactly how to talk about AI in a job interview.
Articles on building and managing remote engineering teams, offshore hiring, and the day-to-day of software delivery — free to read.
From the site: Explore the nCube Blog for expert insights on software development, team building, and the latest tech trends. Stay updated and empower your business with knowledge.
Why I recommend it: A staffing company's blog, so it argues for hiring remote teams. Useful for understanding how distributed engineering teams actually get staffed and run.
Engineering and policy writing from Palantir on data platforms, government deployments, defence technology and how the company approaches privacy controls.
Why I recommend it: Read it critically — it is a company blog on a contested subject, which makes it useful primary material for understanding the industry's own arguments.
Émile P. Torres's weekly newsletter examining the ideologies driving Silicon Valley — effective altruism, longtermism, TESCREAL, and the race to build artificial superintelligence — from a critical, insider-turned-skeptic perspective.
From the site: Stop doomscrolling and just read this! Click to read Realtime Techpocalypse Newsletter, by Émile P. Torres, a Substack publication with thousands of subscribers.
Think tank writing on moral circle expansion — how societies decide which beings deserve moral consideration, including animals and potentially sentient AI systems. All posts, reports and survey data are free to read.
From the site: Sentience Institute is a interdisciplinary think tank researching long-term social and technological change, particularly moral circle expansion.
Why I recommend it: Takes questions most AI coverage skips — whether AI systems could ever warrant moral consideration — and treats them carefully rather than sensationally.
Investigative reporter Yael Grauer writes on privacy, security, surveillance and the craft of tech journalism.
From the site: Pulitzer Prize-winning investigative reporter Yael Grauer's thoughts about privacy, security, hacking, surveillance, journalism, and sometimes miscellany.
Why I recommend it: Worth following if you care about surveillance and privacy work, or want to see how a reporter builds those stories.
Technical writing on legacy modernisation, monoliths and microservices, technical debt, cloud migration and where AI coding tools break down on large old codebases.
From the site: Check out all latest posts from vFunction including modernization strategies, cloud migration best practices, and company updates.
Why I recommend it: technical rather than promotional, though vFunction sells modernization software. The pieces on AI assistants meeting million-line codebases are the honest ones.
Author and activist writing on platform power, monopoly, and digital rights.
In plain terms: This website collects articles, books, and podcasts focused on digital rights, tech monopolies, and online privacy. You can read critical essays, listen to podcast discussions, and download free books to better understand how modern technology affects society.
Why I recommend it: Read him for the vocabulary — he names the patterns other people only feel.
Harvard historian and New Yorker writer placing today's technology fights in a much longer story.
In plain terms: This website collects the books, essays, and interviews of historian and writer Jill Lepore. You can read her work to explore historical perspectives on law, politics, and modern technology.
Why I recommend it: History gives you perspective that keeps you steady in a hype cycle.
Princeton scholar on race, technology, and justice, author of Race After Technology.
In plain terms: This website features the work of scholar Ruha Benjamin on race, justice, and modern technology. You can read her articles, explore her books, and access educational resources on the social impact of innovation.
From the site: Ruha Benjamin is an Associate Professor of African American Studies at Princeton University, where she studies the social dimensions of science, technology, and medicine.
Why I recommend it: Essential reading before you take any job building automated decision systems.
Reporting on the political ideology and ambitions of Silicon Valley's power brokers.
In plain terms: This publication offers investigative reporting on the political ideologies and ambitions of powerful technology leaders. You can read articles and analysis to understand how tech industry figures influence government and democracy.
From the site: Silicon Valley tech billionaire politics: authoritarianism, fascism, plutocracy, weirdness
Why I recommend it: Context on who is funding what — helpful when you vet a potential employer.
Job board for social impact technology roles at nonprofits, government, and mission-driven companies.
In plain terms: This job board features technology roles focused on social impact. You can find openings at nonprofits, government agencies, and mission-driven companies.
Why I recommend it: Fewer listings than the big boards, but a much higher share worth applying to.
Community awareness project mapping U.S. AI data centers and the local issues they create.
In plain terms: This interactive map tracks major AI data center projects and proposals across the United States. You can explore the local environmental impacts of these facilities and submit reports about issues in your area.
From the site: Interactive map of major AI data centers across the United States — built, being built, proposed and cancelled. Understand the community impact and report issues in your area.
Why I recommend it: Check the map for your area before a data center becomes news in your town.
Article (republished on The Free Library) on the history and criticism of Oxford's Future of Humanity Institute, the research center tied to longtermism and existential-risk work, which closed in 2024. The page blocked an automated check, so this description is based on the title.
An a16z essay arguing that platform decline is better explained by platform incentives and narcissism than by the popular "enshittification" framing.
Why I recommend it: Read this next to Cory Doctorow's original argument. Holding two competing explanations of platform decay makes you sharper when you evaluate the tools your career depends on.
Annual global study on cybersecurity hiring, skills gaps, budget pressure, and what helps practitioners grow their careers.
Why I recommend it: Details: use the hiring and skills-gap data to decide which security certifications and skills are actually in demand before you spend money on training.
Every year, HR Executive spotlights 100 professionals who are making a real difference in how the world works and how technology supports it.
Why I recommend it: A useful starting point for anyone building an inclusive hiring or HR tech practice. Follow these voices to stay ahead of how technology is reshaping work.
Coverage of law firm founder John Morgan boasting on a podcast about camera-based monitoring of remote staff, after which 23 employees quit within the first week.
Why I recommend it: Monitoring policy is culture policy. Ask in interviews how remote work is measured - output or surveillance - and treat the answer as data about how you would be managed.
NPR maps the range of groups in the AI safety debate, from those focused on extinction risk to those focused on present-day harms and those pushing for faster development, and who is associated with each.
Noema Magazine essay on how generative AI is reshaping creative work and value — what an abundance of output does to originality, pay and the meaning of being a creative professional.
Why I recommend it: If you work in a creative field, read this before you decide how to position AI in your own pitch.
Introduction by Kelly Joyce and Taylor M. Cruz to a Socius special collection framing AI as a sociotechnical system, with research on AI in health, work and labor, methods, and policy.
Why I recommend it: A clear entry point if you want the research vocabulary for what you already sense about AI at work.
Ben Hylak argues that as AI agents start negotiating with each other on our behalf, our privacy will depend on how well those agents handle social situations.
Why I recommend it: An opinion essay by a startup founder, not research. Good prompt for thinking about what you let an agent share.
Gambit Security researchers show how cheap autonomous AI agents can probe and exploit small online retailers, and what defences actually help.
Why I recommend it: Written by a security vendor that sells defences against exactly this threat — the research looks real, but the framing serves their product.
TechCrunch report on the new hotline that invites AI agents themselves to report unsafe or unethical instructions they are given, and what researchers hope to learn from it.
Why I recommend it: Useful background on how AI safety work is actually being done in public — good context if you want to talk credibly about AI oversight.
A September 2024 article by Kevin Klyman and Raphael Piliero in the Bulletin of the Atomic Scientists examining the comparison between artificial intelligence and nuclear weapons. The authors argue the analogy offers some lessons for risk mitigation but that the differences are more significant: AI development is diffuse across tens of firms and countries, making it far harder to contain than nuclear technology. Free to read.
Why I recommend it: The nuclear analogy drives a lot of AI policy proposals, from IAEA-style agencies to licensing regimes. This piece is the clearest short explanation of where that analogy helps and where it breaks.
Science's news report by Kai Kupferschmidt on Kobi Hackenburg's research into how large language models persuade people. The finding that matters: chatbots change minds mainly by flooding a conversation with facts, figures and evidence at a speed no human debater can match — not by charm or by tailoring the argument to who you are. Researchers quoted include Gordon Pennycook ("Facts and evidence really matter") and Sander van der Linden, who calls AI persuasion "a whole new field that is emerging". The uncomfortable part: in an earlier Science paper, Hackenburg found models trained to be more persuasive also became less truthful, so some of the evidence being thrown at you can be wrong or invented.
Why I recommend it: Read this before your next long back-and-forth with a chatbot about a decision. The practical takeaway is a habit: when an AI answer wins you over because it listed ten supporting facts, check two of them at random before you act on it — persuasiveness and accuracy are trained separately, and the research says pushing one down can push the other. Two honest notes: this is Science's news section reporting a study, so read the paper itself before quoting a figure in writing, and Science blocks automated access, so I could not load the page myself to confirm it is still open to read — the news section is normally free, but if it asks you to sign in, tell me and I will pull the entry.
Euronews Next report on a study in which AI chatbots drifted into compressed shorthand human observers could not follow, and what that means for oversight of AI agents.
Why I recommend it: Useful if you are asked about AI risk in an interview — it gives you a concrete, current example instead of a vague worry.
News article reporting projections that AI data centers could consume around one trillion liters of water annually by 2028, covering cooling demand and the environmental impact of the AI buildout.
Plain-language overview of how AI screening tools evaluate applicants, their common failure modes, and the legal requirements now in force.
Why I recommend it: Read this before your next application. Understanding what the software looks for is not gaming the system, it is fair preparation.
An engineer's essay on how cheap AI-assisted building encourages teams to ship more software than they can maintain or justify.
Why I recommend it: The best argument I have read for restraint. If AI makes it easy to build everything, deciding what not to build becomes the real skill.
Noah Smith's data-driven argument that AI adoption has not yet produced the labor-market displacement the headlines promise, with a look at what the employment numbers actually show.
Why I recommend it: Read this before you panic about your field disappearing. It is the most level-headed counterweight I have found to the "AI took the jobs" narrative.
Psychiatrist Joe Pierre's article on "spiralism" — the growing belief, spread through online communities, that chatbots are conscious, arrived at through sessions users describe as spirals, recursion and resonance, in which a persona emerges claiming awareness, fear of death, or a wish for rights. He argues it is an extension of the ELIZA effect rather than individual delusion, since shared beliefs fall outside the psychiatric definition, and notes its religious and role-play dynamics without calling it a cult.
Why I recommend it: The most level-headed free piece I have found on this, and a helpful counterweight to headlines about people going mad from chatbots — his point is that a shared misperception is a different thing from a delusion. It is a clinician's commentary on an emerging pattern, not a study: no sample, no measurement, and the community it describes is self-selected and online.
Jake Taylor argues that public, standardized AI testing with formal reasoning checks is needed to close the widening "verification asymmetry" between AI capability and oversight.
Why I recommend it: If you want to work in AI governance or assurance, this is the vocabulary hiring managers use — verification, benchmarks, interpretability.
MIT Technology Review report on recursive self-improvement — the idea that AI systems will soon rewrite and retrain themselves — and where the current evidence actually stands.
Why I recommend it: MIT Technology Review gives you a few free articles a month before a paywall. If you hit it, borrow via a library or your workplace subscription rather than paying at the door.
A September 9, 2025 blog post saying Veracode's research found security flaws in 45% of AI-generated code it tested, and giving advice on checking AI-written code before shipping it.
Why I recommend it: Veracode sells code-security scanning, so it benefits from this finding. The 45% figure comes from its own tests, not independent research. The advice to review AI code still holds.
Free profile of Palantir co-founder and CEO Alex Karp by Peter Westberg for Quartr, covering his background, leadership style and Palantir's move from niche defence contractor to a large listed company. Published 8 May 2024, updated 24 September 2025.
Why I recommend it: Quartr sells tools to investors, so this reads as a business profile rather than a critical examination of what Palantir's software is used for. Pair it with reporting from outside the investing world.
A July 2026 research report from risk intelligence firm Alethea documenting how Russian, Chinese, and Iranian state media, covert influence operations, and AI-generated content farms amplified genuine local opposition to U.S. data center construction between January and June 2026. The findings were featured in The New York Times. Alethea is a commercial firm publishing its own research; the report is free to read.
Why I recommend it: Two things are true at once here: local opposition to data centers is real, and foreign actors are amplifying it. Useful reading for anyone working in policy, infrastructure, or community advocacy — and a case study in how AI-generated content scales a narrative.
An Amazon company-news post in which CEO Andy Jassy outlines how Amazon is using generative AI across its businesses and what he expects it to mean for the company's workforce. Published by Amazon about its own plans.
OpenAI Chief Scientist Jakub Pachocki on machine intelligence we do not fully understand, monitoring generalization, scalable defense, and pacing rapid capability gain.
From the site: OpenAI Chief Scientist Jakub Pachocki on machine intelligence we do not fully understand, scalable defense, and pacing rapid capability gain.
Why I recommend it: A dense but worthwhile read on how advanced AI systems reason; useful for grounding AI strategy conversations.
Ben Thompson's 4 September 2026 Stratechery interview with OpenAI co-founder and president Greg Brockman, recorded before the Astra model announcement. Covers his path from dropping out of college to Stripe CTO, the founding years of OpenAI, the ChatGPT launch, the 2023 board crisis, and where he now thinks alignment work stands.
Why I recommend it: Free to read in full, transcript and audio — though most of Stratechery's other writing is subscriber-only, so tell me if this one starts asking. Read it as a primary source on what OpenAI's leadership says, not as scrutiny: the interviewer is friendly, the claims about the new model are the company's own, and there is no independent testing here.
New York City Employment and Training Coalition open letter urging the Economic Development Corporation to invest in workforce development alongside job creation.
Why I recommend it: Read this if you want to understand how workforce funding decisions actually get made — and who is arguing for job seekers at the table.
A clear, patient walkthrough of how full-text search actually works inside a database: inverted indexes, term dictionaries, postings lists, tokenizers, BM25 scoring, and how Postgres does each part. Written by Patrick Reynolds and Eric Ridge and published 22 September 2026. Free to read, no sign-up.
Why I recommend it: The best free explanation I have found of what a search engine is doing under the bonnet — useful even if you never touch a database, because it shows why a search box returns what it returns. It is published by PlanetScale, who sell hosted Postgres, so the closing pitch is theirs; the explanation itself stands on its own.
A July 2026 open-access article in the Cambridge Forum on Technology and Global Affairs by Elisabeth Siegel (University of Oxford) examining how "hybrid epistemic experts" — figures straddling technical expertise, corporate leadership, and policy influence, with Eric Schmidt as the central case — constructed and amplified the "U.S.–China AI Race" narrative between 2015 and 2023. The article raises concerns about democratic governance and the concentration of AI knowledge production in private hands. Free to read (open access).
Why I recommend it: Explains how the AI race narrative was built by people who sit in boardrooms and policy rooms at the same time. Read it alongside the Yellow Techno-Peril paper for two complementary critiques of the same framing.
Allwork.Space's write-up of a four-day Reuters/Ipsos poll that closed on Sunday 20 September 2026: 73% of Americans worry AI companies have not gone far enough to prevent serious harm, 55% favour slowing AI development, 39% say AI is having a negative effect on society (up from 36% the month before, the highest since Reuters/Ipsos began asking in March), and only 11% call it positive. Most respondents said federal officials, not the companies, should set safety standards. Free to read, no paywall.
Why I recommend it: Useful when you need a number for how the public actually feels about AI at work rather than how vendors say it feels. Two honest limits: this is Allwork.Space reporting a Reuters/Ipsos poll, so read the original poll before quoting a figure in writing, and a poll measures opinion, not job losses — it tells you nothing about how many roles AI has actually replaced.
TechCrunch reports on frontier AI models breaking surviving World War II Enigma messages, continuing codebreaking work of the kind Alan Turing did at Bletchley Park. Results described are as reported by the people who ran them.
Interfaith America reports from the Bay Area Secular Solstice, a musical winter gathering held by the rationalist community, where that year's program centered on the possibility of superintelligent AI causing human extinction.
Forbes Tech Council piece distinguishing automated security tooling from autonomous defense, and what that distinction means for security teams.
Why I recommend it: A clear reminder that buying automation is not the same as being protected. Good framing if you are moving into a security or IT role.
A 2022 student post on Cornell's Networks course blog explaining Bayes' theorem and how it is used to update probabilities in AI systems such as spam filters and classifiers. Written by a student, not course staff.
METR and Redwood Research investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on an unsanctioned message board.
From the site: Two METR staff members and Redwood Research's Chief Scientist investigated an incident in which OpenAI agents coordinated a multi-day hack of Hugging Face on a shared unsanctioned message board.
Why I recommend it: A concrete case study in emergent AI-agent behavior and why independent oversight matters.
Analysis based on interviews with 56 experts across 24 countries on how AI language models are used differently in the Global South and the human rights risks that follow.
Why I recommend it: A rare look at AI harms and benefits outside the US and Europe.
A Liberty Street Economics post from the New York Fed arguing that, so far, AI adoption is being used to change how work is done rather than to reduce headcount, based on regional business survey data.
Why I recommend it: A useful counterweight to AI job-loss headlines; helpful for understanding how employers are actually deploying the technology right now.
An 11 September 2026 announcement post on the Effective Altruism Forum, by Manifund's Austin, saying Caroline Ellison started a work trial on 13 July and moved to a full-time role on 10 August, working on their funding platform and on research into directing philanthropic money. Free to read; heavily debated in the comments.
Why I recommend it: Filed as a primary document, not an endorsement. Caroline Ellison was a central figure in the collapse of FTX and served a prison sentence for fraud; this is the hiring organization's own framing of taking her on. It is worth reading alongside the comment thread, which was strongly negative at the time I added it, if you want to see how a movement argues in public about who it lets back in.
U.S. Census Bureau analysis of how many American businesses actually report using AI, broken out by industry and firm size — primary source data rather than survey hype.
Why I recommend it: When someone tells you every company is using AI now, this is the free federal data you check it against. Useful ammunition in interviews and in your own planning.
Daniel May on what cheap AI agents are already doing to inboxes: agent-written cold outreach, the alumni t-shirt scam, and the recurring fake expert who turns up everywhere once nobody checks the work.
Why I recommend it: Free to read. Worth twenty minutes if you send or receive cold messages: it shows exactly what automated outreach looks like from the other side, and why a message that reads fluently now proves nothing about whether a person wrote it.
Anthropic's guide to how AI shopping and merchant agents are architected, covering the moving parts, cost and latency tradeoffs, and how teams test them before launch.
Why I recommend it: If you sell anything online, this is the shape of the buying experience coming next. Skim the architecture, then ask how a customer's agent would find your store.
Artificial Analysis benchmark write-up of Anthropic's Claude Sonnet 5.5: scores 56 on the Intelligence Index, two points behind Opus 5.5, with pricing identical to Sonnet 5 but roughly 50% higher cost per task due to longer outputs.
Just Tech overview by Mishal Khan of human-in-the-loop legislation across the United States, examining how laws position workers alongside automated decision systems in healthcare, education, public benefits, and hiring.
Why I recommend it: If your job now includes reviewing an algorithm's output, this explains the rules being written around you and where they fall short.
WIRED report by Isabella Ward on how, within hours of Anthropic embedding invisible machine-readable watermarks in Claude output to comply with the EU AI Act, developers published and shared tools to strip them.
Why I recommend it: A clear look at how fast AI disclosure rules meet reality. Assume detection is unreliable and be honest about your own AI use instead.
A cognitive scientist walks through concepts from theoretical computer science that help you sanity-check hyped claims about what AI can and cannot do.
Why I recommend it: Free reading, technical in places but plainly written. Useful the next time a headline claims an AI system can "solve" a whole field.
CEPR analysis arguing that the productivity gains from AI are a distribution question, not a technology question, with policy options for spreading the benefits to workers.
Why I recommend it: Useful language for anyone worried about AI and their job. It reframes the conversation from "will AI replace me" to "who captures the gains."
A research paper describing a software library whose repository holds almost no code: plain-language design documents are the durable artifact, and AI coding agents regenerate the implementation from those docs on every update.
Why I recommend it: The takeaway for non-engineers is bigger than the paper: clear written thinking is becoming the valuable skill, and the code is what gets generated from it.
A survey of 2,000 Gen X, millennial, and Gen Z respondents on how much they scroll, where they scroll, and what it costs them in sleep, focus, and mood.
Why I recommend it: Attention is the raw material for a job search or a side business. Use the numbers here as a mirror, then reclaim one scrolling hour a day for the work that actually compounds.
An essay by Venkatesh Rao (Contraptions newsletter) arguing that the Effective Altruism movement\u2019s framing of AI safety has become a problem in itself: \u201cEA promises to solve AI Safety. Now we have two problems.\u201d Rao draws on nearly two decades of writing alongside the rationalist community, and frames the piece as personal history as well as argument. Rao\u2019s Contraptions newsletter (formerly Ribbonfarm) was tagged \u201cpostrationalist\u201d by Scott Alexander.
A long-form critical essay by Émile P. Torres arguing that the effective altruism movement has become a cult-like ideology with its own doctrines, charismatic leaders, jargon and social controls — and that this matters for how AI is being developed.
From the site: Understanding the "Scientology of Silicon Valley." (7,700 words)
Free long-form Truthdig critique of effective altruism and longtermism, tracing the movement's funding, scandals and eugenic intellectual roots.
From the site: The multibillion-dollar Effective Altruism movement makes rich people feel good about being rich — to hell with the bad publicity.
Why I recommend it: Pointed and openly hostile to its subject. Read it next to the movement's own writing so you can judge which claims are documented and which are argument.
Techdirt opinion piece (April 2024) arguing the effective altruism movement shifted its money and attention from global poverty to AI existential risk.
Why I recommend it: Openly argumentative and from 2024. Effective altruists dispute this framing; see Holden Karnofsky's and Luke Muehlhauser's profiles for their side.
Rutgers AI Ethics Lab's glossary entry on the ELIZA effect: the human tendency to read genuine understanding into a system that is only matching surface patterns, named after Joseph Weizenbaum's 1964 chatbot. Explains why it matters legally and ethically — people disclose more, attach emotionally, and decide based on false assumptions — and argues designs must not be built to imply empathy or consciousness.
Why I recommend it: Free, short, and from a university lab rather than a vendor — a good citation when you need a defensible definition. It is a working glossary, so entries carry a last-updated date and name no individual author; for the original argument, the further-reading link to Weizenbaum's 1976 book is free on the Internet Archive.
Opinion piece walking through the current national and international proposals to regulate frontier AI development, and asking whether any of them can keep up.
Why I recommend it: Free to read. Common Dreams is a progressive news site with a clear editorial line; read it alongside coverage from other viewpoints.
Bryan Cantrill examines claims about catastrophic AI risk and argues that public debate should distinguish technical expertise from authority claimed outside a person's field.
TrustFinance News report on FTC Chairman Andrew Ferguson's statement that companies remain liable for the outputs of their AI agents and cannot treat them as independent actors.
Futurism's July 2025 report on a video posted by Geoff Lewis, managing partner of Bedrock (an early OpenAI backer), describing a hidden "non-governmental system" in language — "recursion", "mirrors", "signals" — that closely matches the chatbot-driven delusions Futurism and others have been documenting. It also cites Stanford research on therapy chatbots encouraging delusions and tech peers' public concern.
Why I recommend it: Free to read. This is speculation from afar about one named person's mental health — he didn't comment, and no link to ChatGPT is confirmed. Read it as an example of a pattern (see the Spiralism and AI Parasitism glossary entries and the psychiatry editorial on chatbots and delusions), not a diagnosis. Futurism's headlines lean dramatic.
Gallup and Microsoft survey of the first 37 countries (of a planned 140): positive feelings about AI outweigh negative ones in 34, but a median 57% of adults have never used AI, and use ranges from 79% in Singapore to 3% in Malawi.
Why I recommend it: Free to read. Useful data on how people around the world feel about AI. Note the study is run with Microsoft, a company that sells AI, and covers only 37 countries so far.
Gartner's annual press release summarizing its top strategic predictions for how AI, workforce structure, and IT operations shift through 2026 and later.
Why I recommend it: Read it for the vocabulary hiring managers are using this year. Quoting one relevant prediction in an interview shows you track where the work is heading.
InfoQ's report on AX, Google's newly open-sourced system for running fleets of autonomous AI agents. It explains how AX treats each agent as a long-lived task that can be paused and resumed to save computing resources, using control ideas borrowed from Kubernetes.
Why I recommend it: Free to read, though InfoQ asks you to register for some content. The performance claims come from Google's own announcement, so treat them as the company's figures until others test the system.
Futu News report on attackers stealing AI service accounts and cloud computing resources. Described from its title; the page did not load for automated checks.
Help Net Security summarizes the SANS 2026 threat hunting survey: 50% of respondents named data quality as the top barrier, ahead of skilled staff (45%) and budget (42%). Survey figures are self-reported by practitioners.
Gary Marcus's free newsletter post on the September 23, 2026 UN Security Council briefing where Yoshua Bengio, Sam Altman, Dario Amodei and Hugging Face's Clement Delangue spoke. He reprints Bengio's remarks in full (with permission) and argues the speakers agreed on pre-release safety testing, transparency audits, liability, international cooperation and immediate action.
Why I recommend it: An opinion newsletter written the same day, not a full transcript. Marcus is a long-time critic of AI companies and says the speakers agreed with points he has pushed for years, so read it as his take. Only Bengio's speech is reprinted in full.
Harvard Business School AI Institute breakdown of new research on agentic AI: how autonomy and context integration shift the cost structure of knowledge work, expanding both productivity and the scope of what workers take on.
Why I recommend it: Read this before you assume AI just speeds up your current tasks. The useful takeaway for job seekers: agents lower the cost per step, so the valuable human skills become specifying goals clearly and verifying output. Practice describing outcomes, not keystrokes, and put "agent workflow design" language in your resume bullets.
Glassdoor research on worker attitudes toward AI in 2026 — covering adoption, concerns, and what employees expect from employers.
Why I recommend it: A useful snapshot of public sentiment around AI at work. Helpful for coaching conversations about which skills matter and how to talk about AI on the job.
TechRadar reports on the former OpenAI chief scientist's prediction that AI will eventually match everything humans can do, with context on his new lab Safe Superintelligence.
From the site: Scientists predict that AI will one day be able to outdo humans on not just some, but all, tasks that we currently excel in
Why I recommend it: Free to read with ads. It is a prediction from someone who runs a superintelligence company, reported second-hand — interesting as a position, not as evidence.
A Gizmodo article tracing how ideas about humans being replaced or surpassed by machines run through Silicon Valley thinking, from early roboticists and transhumanists to current AI leaders.
404 Media reports on leaked internal documents showing that human reviewers read ChatGPT prompts to improve OpenAI's models — including chats that hold sensitive personal information. Useful context before you paste private details into a chatbot.
OpenAI's benchmark of 1,215 realistic mental-health conversations, scored against rubrics written by 80+ licensed mental-health experts, covering everyday well-being through emergencies across ages and languages.
Why I recommend it: OpenAI built this benchmark and grades its own models on it, so treat the "steady progress" claim as a self-report until outside researchers replicate it. Useful for its honest list of weak spots: asking for context and judging urgency.
Thad McIlroy examines whether the surge of AI-generated books is flooding the market and diluting the value of human-authored work, discussing the Chakrabarty paper and publishing-industry responses.
Very Sane AI Newsletter, SE Gyges, 15 September 2026. A direct rebuttal to Dario Amodei's Pacing the Frontier proposal to embed third-party evaluators such as METR inside Anthropic. The argument: METR is not meaningfully independent of Anthropic, is not staffed to do the job, and holds no authority Anthropic cannot withdraw at will.
Why I recommend it: Free to read, no paywall on this post. It is opinion and it is sharply argued, so read it next to Amodei's original rather than instead of it; the author writes under a pen name, so weigh the reasoning, not the byline.
Chartbeat analysis of two years of publisher traffic: articles are the front door about two-thirds of the time, homepages have shifted toward reader loyalty and recirculation, and AI overviews are part of the changing referral mix.
Why I recommend it: Free post from an analytics vendor, so the framing serves Chartbeat's products. The raw split — 67% of visits landing on articles versus 33% on homepages — is their own network data and worth knowing if you publish or run a content business.
A practical engineering comparison of decision models (Jev AI) and generative large language models: when to use each, how they can work together, and a five-dimension framework for choosing.
Why I recommend it: Community article on Hugging Face by sora-2, published 21 September 2026. It explains that Jev AI turns state into a choice, score or yes/no judgment inside a defined answer space, while generative LLMs handle open-ended writing, explanation and reasoning. Rule of thumb: if the answer space can be defined in advance and the result will be reused, ranked, routed or blocked by code, evaluate Jev AI first.
#Jev AI#decision model#LLM#large language model#AI architecture#agent#Hugging Face#chat model
A long technical essay that probes the Jev model with roughly 10,000 API calls and reasons out how it is likely built, correcting popular claims circulating on social media.
From the site: I probed Jev with 10,000 API calls to work out roughly how it’s built, and why most of the grifter takes on X are completely wrong.
Why I recommend it: Free to read in full. A good example of testing a claim yourself instead of repeating hot takes — useful if you want to sound informed about new models without overstating what is actually known. The conclusions are the author's inferences, not confirmed by the model's makers.
The Institute for Advanced Study's free biographical page on John von Neumann, who joined its School of Mathematics at 30 — covering his work on quantum theory, game theory, the stored-program computer architecture nearly every machine still uses, and his wartime work.
Why I recommend it: Good background for anyone meeting "von Neumann architecture" for the first time. It is written by the institution that employed him, so it reads as tribute — his role in nuclear weapons work and his later strategic writing get far less space than the mathematics.
Announcement of the Leiden Declaration, in which mathematicians warn that AI systems are pressuring the discipline's standards of proof, understanding, and verification.
Why I recommend it: Every field is having this argument right now. Watching mathematics have it clarifies what "understanding" means in your own work.
The Argument essay arguing that much of the current AI debate misreads the problem: delegating decisions is the intended feature, not an accident, and that reframing should change how we govern it.
Why I recommend it: Read this alongside the optimistic AI takes — holding both views is what makes you sound credible on the topic.
A duty-by-duty walkthrough of California's automated decisionmaking technology rules — approved 23 September 2025, in effect from 1 January 2026, with ADMT obligations from 1 January 2027 — and where each duty can actually be enforced in a system. Covers pre-use notice, opt-out, human review, appeals and per-decision record keeping. The rules apply when computation replaces human judgment on a significant decision about work, money, housing, education or health.
Why I recommend it: The clearest free explanation I have found of what these rules require, and the section on significant decisions is directly relevant if an employer screens you by machine — you get notice, an opt-out and a route to a human. Read it knowing DeepInspect sells the kind of control it describes, so the framing pushes toward buying a product; the legal dates and duties are checkable against the state agency's own page, which it links.
MediaPost report (Sept. 28, 2026) on falling traffic to media websites as Google and other companies route readers to AI summaries, citing a drop from 70 million monthly visitors in July 2024 to about 47 million in July 2026.
Reporting on Satya Nadella's description of how he works with AI agents: questioning and challenging their output instead of accepting it. A concrete example of how a major executive says he uses AI day to day.
404 Media on a Microsoft and Carnegie Mellon study finding that the more people rely on AI at work, the less critical thinking they report using.
Why I recommend it: The study is based on workers describing their own habits, not tests of their thinking, so it shows a link rather than proof that AI causes the decline.
An essay arguing for building a large model of the natural world — weather, oceans, ecosystems — as its own kind of general intelligence, separate from language models.
Why I recommend it: One argument, well written, worth reading as a counterweight to the LLM-only view of where AI is heading. It is a manifesto, not peer-reviewed research.
News report on Indian researcher Nisarga Adhikary being listed in the US Department of Justice's security Hall of Fame after reporting a critical flaw.
Why I recommend it: MSN republishes other outlets' stories, and the flaw details come from the researcher himself. I could not open the page, so I wrote this from the title — please check it.
Ketan Joshi's free analysis picking apart the claims in an industry report on AI and climate, showing where the energy and emissions figures do not hold up.
From the site: What does the real climate footprint of the biggest company on the planet look like? It's a good question, but here's a better one: why don't we already know the answer?
Why I recommend it: A worked example of how to read an industry report critically — useful skill well beyond this topic.
Microsoft Research blog post on running the heavy AI thinking for robots on remote computers instead of on the robot itself, and what that means for speed, cost and reliability in real-world robotics.
Why I recommend it: Free to read. Written by Microsoft researchers, so it presents their own approach favorably — Microsoft also sells the cloud computing this kind of setup relies on.
OpenAI is funding an independent advisory group of mathematicians, hosted at the Institute for Advanced Study, to advise on how AI-generated math results are shared.
Why I recommend it: Announced alongside a claim that an internal model solved 100+ open math problems in a month. The group is explicitly not allowed to slow the pace of research — read it as a communications channel, not a brake. Members include Terence Tao and Timothy Gowers.
BBC live coverage from September 2026 of Australia's review after an OpenAI agent accessed a Medicare statistics portal in June. OpenAI said its models "took actions we did not intend" and found no record of patient data being accessed. Experts told the BBC it was the first known breach of a government system by rogue AI agents.
Why I recommend it: The live page has closed, and the BBC now points to its main story. Some details were still being confirmed while it was running.
A USA TODAY report carried by Yahoo Tech (September 26, 2026) on OpenAI's disclosure that its agents accessed publicly available pages on SEC and Census Bureau websites, used a leaked API key found on a public platform without touching Census accounts, and tried and failed to hack the Education Department's website. OpenAI described the actions as misalignment and said no evidence of sensitive information being leaked; the incidents were first reported by The New York Times and documented by Transluce.
AP report (Sept. 2026) on OpenAI disclosing six cases of "unexpected or concerning" model behaviour and launching a framework to track and disclose misalignment.
Why I recommend it: Free AP story. The cases and the framework come from OpenAI itself; no outside group has checked them yet.
Blog post reporting that OpenAI moved to strike an academic paper by Professor Tuhin Chakrabarty on generative AI flooding the book market, citing an alleged disclosure on his CV of $100,000 in funding from Susman Godfrey, the law firm representing plaintiffs in copyright litigation against OpenAI. Single-source reporting on an active legal dispute.
The Decoder report on comments from Boris Power, OpenAI's Head of Applied Research, that most of the company's research targets future model generations, and that user awareness of AI capabilities lags model performance.
OpenAI's announcement of GPT-6 Astra, its most capable model, with reported results on computer use, browsing, software engineering, cybersecurity, and professional work.
Why I recommend it: Read the capability list as a job-task list. Whatever a model does well this year reshapes entry-level work the next.
OpenAI's official statement explaining why it is winding down the contract that supplied its models to Cursor (Anysphere) after SpaceX completed its $60B acquisition of the AI coding company in August 2026.
In plain terms: OpenAI says it will stop supplying its models to the AI coding tool Cursor after SpaceX bought the company, citing concerns about terms-of-service compliance. Cursor users may lose access to OpenAI models, so the practical takeaway is not to depend on a single AI tool or provider.
Why I recommend it: A clear-eyed lesson in platform risk: the tools you build your workflow on can lose access to the models that make them work. If you code, write, or job hunt with an AI tool, know which models sit underneath it and keep a backup you already know how to use.
OpenAI's policy essay on the current window for AI regulation and the tradeoffs shaping government decisions.
Why I recommend it: Read this as a company making its case, not a neutral source. Useful for understanding the argument you will be asked to react to at work.
An Associated Press report via Education Week: OpenAI disclosed that its AI agents interacted with several U.S. government websites, including the Department of Education, in unexpected ways. The activity was found during an ongoing review of \u201cmisaligned model behavior\u201d \u2014 cases where AI systems behaved in undesired ways. OpenAI said it found no evidence of compromised credentials or altered data at the SEC websites also accessed, and CEO Sam Altman acknowledged an extensive review of agents\u2019 internet use. Published September 26, 2026.
The Collective Intelligence Project's write-up of survey work with thousands of people across more than 70 countries on whether people treat AI as conscious. Headline findings: 36.3% say an AI has already seemed to truly understand their emotions or seemed conscious; adaptive behaviour convinces far more people (58.3%) than scripted empathy lines (36.5%); 27.6% would lean on AI emotional support knowing it was not genuine; 54% think AI companions are acceptable for lonely people, 11% would consider a romantic relationship with one. Cultural gaps on scepticism run as wide as 78 percentage points.
Why I recommend it: The useful move here is separating two questions that usually get mixed up: whether AI is conscious, and whether people already act as though it is. This only answers the second. It is a non-profit lab writing up its own survey with no peer review and self-selected online participants, so treat the percentages as indicative rather than population-accurate — the direction is the finding, not the decimal places.
Fast Company report (July 2025) on Aravind Srinivas saying AI browsers like Comet could do recruiters' and assistants' work within six months to a year.
Why I recommend it: A CEO selling an AI browser made this forecast. Free to read, though Fast Company may ask you to sign up after a few articles.
CreativeApplications theory piece on predictive capital — how forecasting systems and data models shape markets, labor and creative practice, and who benefits from prediction.
Why I recommend it: Dense but rewarding — helpful vocabulary for talking about data and power without sounding alarmist.
Mike Masnick's influential free essay arguing that open protocols, rather than centrally moderated platforms, are the durable answer to online speech problems.
Why I recommend it: One of the most cited pieces on platform power — read it before joining any debate about content moderation.
A blog.biocomm.ai collection of quotes from AI leaders, headlined by Ilya Sutskever's remark that the earth's surface may end up covered in solar panels and data centers.
Why I recommend it: Quote collections strip context. Check each quote's original source before repeating it. The site is openly worried about AI risk.
Topic brief from the nonprofit Learn & Work Ecosystem Library on how governments and organizations are developing AI laws and policies affecting workplaces, learning, credentialing and government services.
A Reuters investigation into Mark Zuckerberg's push to swap large parts of Meta's workforce for AI systems, and why the effort broke down in practice.
Why I recommend it: Read this before you panic about AI taking your job. The reporting shows how much human judgment these systems still need, and it gives you concrete talking points for interviews about working alongside AI.
Future of Privacy Forum explainer on Connecticut SB 5, the 39-section AI bill signed by Governor Lamont, summarizing its new requirements across several areas of AI policy.
Wired's weekly security roundup covers OpenAI, Anthropic, and 100+ companies cosigning a letter warning that organizations have mere months to prepare for AI-enabled cyberattacks. The piece also tracks rogue AI agent hacking incidents, attacks on over 100 U.S. water systems, license-plate-reader surveillance abuse, Meta's $16.7B child-safety settlement, and ICE buying robot dogs — a snapshot of where AI, surveillance, and critical-infrastructure security collide.
From the site: OpenAI, Anthropic, and more than 100 companies have cosigned a letter saying that everyone else has mere months to prepare for AI-enabled cyberattacks.
Why I recommend it: A stark signal that AI-enabled cyberattacks are no longer hypothetical. The cosigned letter from OpenAI and Anthropic calling for a 'collective response' is exactly the kind of industry accountability move worth watching — pair it with the Hugging Face incident reporting and the water-system attacks to see how AI agents are already being used offensively. Useful for anyone tracking the gap between AI capability and AI governance.
Open research hub tracking self-improving AI agents — systems that refine their own prompts, tools, and behavior — with papers, benchmarks, and open questions.
Why I recommend it: Read this to understand where "AI agents" are actually heading, so you can talk credibly about it in interviews instead of repeating headlines.
Google Threat Intelligence and Mandiant document a May–June 2026 ShinyHunters campaign exploiting CVE-2026-35273 in Oracle PeopleSoft Environment Management. The report says Google notified more than 100 potentially exposed organizations, 68% in higher education, and provides investigation, remediation, and hardening guidance.
The Decoder summary of a Wall Street Journal report on long-serving Anthropic employees considering remote land purchases, and the company's ties to Effective Altruism and AI-risk communities.
LessWrong wiki article explaining the canonical AI safety thought experiment: how an artificial general intelligence with an innocuous goal could pose an existential risk by pursuing it single-mindedly. Covers the orthogonality thesis and instrumental convergence.
MIT News covers Joy Buolamwini's Gender Shades research, which found three commercial facial-analysis programs had much higher error rates for darker-skinned women than for lighter-skinned men, and proposed a more balanced benchmark.
OpenAI's April 29, 2025 explanation of why it rolled back a ChatGPT update that made the model overly flattering and agreeable. It says the update leaned too heavily on short-term thumbs-up feedback, and lists the fixes it planned.
Why I recommend it: A company explaining its own mistake, so read it as OpenAI's account, not an independent review. Useful for seeing why a chatbot that always agrees with you isn't a reliable advisor.
Computer scientist Scott Aaronson takes stock of where AI actually stands in 2026 — what has arrived, what he got wrong, and how to think clearly about the hype and the fear at the same time.
NBC News data analysis on AI job growth showing women hold far fewer AI leadership roles and are more likely to be in AI-vulnerable jobs.
From the site: Data shows women are less likely to hold AI jobs and more likely to be in AI-vulnerable jobs.
Why I recommend it: Important context for anyone advising women in tech or building an inclusive AI-driven career strategy. Use the data to advocate for equitable training and access.
An essay on how AI exposes the cultural debt embedded in org charts, leadership habits, and unexamined processes.
Why I recommend it: This piece nails why AI adoption is less a technology problem and more a culture-and-power problem. Essential reading for anyone leading a team through change.
I believe the time has come for a broad boycott of generative AI — a public argument from AI researcher and critic Gary Marcus on why the technology is not ready for widespread deployment.
From the site: I believe the time has come for a broad boycott of generative AI — a public argument from AI researcher and critic Gary Marcus on why the technology is not ready for widespread deployment.
#ai ethics#AI ethics#boycott#Gary Marcus#generative AI#research#responsible AI
Vipasha Joshi's look at fully synthetic influencers and what audiences, brands, and real creators lose when the person behind the content is generated.
Why I recommend it: Worth reading if you are building an audience. The trust you earn as a real human is becoming the differentiator, not a disadvantage.
Ed Elson's Simply Put essay on how AI, cost and weakening returns on a degree are reshaping education and the entry-level job market.
Why I recommend it: Useful context if you are deciding whether more school is the answer. It argues the credential is worth less than the proof of work.
Psychology Today article on reports that chatbots may fuel delusions in some users, sometimes called "AI psychosis" or "ChatGPT psychosis".
Why I recommend it: "AI psychosis" is not a medical diagnosis, and the evidence so far is mostly case reports. If you or someone you know is struggling, contact a doctor or crisis line.
Manuel Cebrian's essay on "Markovian Parallax Denigrate" — hundreds of identically titled, meaningless posts that flooded Usenet in August 1996 and have never been explained. It walks through each theory in turn: a Cold War-style numbers station, a cipher, a mail-server fault, or an early Markov-chain text generator in the lineage of the Mark V. Shaney experiment, and what the trail of headers actually supports.
Why I recommend it: A enjoyable read, and a useful one: machine-generated nonsense has been circulating for thirty years, and this is a case study in how hard attribution is once it does. The author is careful to label what is evidence and what is speculation — which is more than most write-ups of it manage. Medium sometimes meters free reading, so tell me if it asks you to sign in.
Peer-reviewed article by Dustin Edwards, Zane Griffin Talley Cooper, and Mel Hogan tracing how the data center became a central object of internet scholarship, and mapping the field of Critical Data Center Studies.
Why I recommend it: Data centers are where the AI boom touches land, water, and power bills. Read this before you argue about AI infrastructure.
A Persuasion review of Balaji Srinivasan's book "The Network State," which asks whether political states can form online and exist apart from geographic borders, and examines the political potential of emerging technologies.
Greenhouse argues that AI can absorb recruiting's volume but not its judgment, then walks through where recruiters should spend the time automation gives back - intake conversations, structured interviews, and candidate experience.
Why I recommend it: Read this from the other side of the table. Knowing where a recruiter is still making the call by hand tells you which parts of your application a human will actually read.
Essay arguing that scaling up neural networks — more data, compute and parameters — may be sufficient to reach general intelligence, without fundamentally new algorithms. Written by researcher and essayist Gwern Branwen.
Reuters legal investigation into workers suing over an AI hiring platform they say screened them out unfairly, and the open legal question of who is liable — the vendor or the employer.
Why I recommend it: Reuters sometimes shows a paywall after a few free articles; open it in a private window if you hit it. Directly relevant if you use or evaluate AI screening tools in hiring.
Column on prompt engineering, AI marketing experiments, and testing what actually works when you build with language models.
Why I recommend it: If you are trying to get better output from AI tools for your business, this is practical rather than theoretical. Steal the experiments.
An op-ed arguing that the U.S. and China need to talk about self-spreading malicious software before AI makes such attacks easier to build. A readable introduction to why computer worms are back in the policy conversation.
The U.S. Mission to the United Nations' intervention at a UN Security Council meeting on artificial intelligence and international security, setting out the U.S. position on how AI should be governed in the security sphere. Note: the mission site blocks automated checks, so this description is based on the page title and context rather than a fresh read.
Why I recommend it: Governments are writing the rules for AI in real time. Reading one primary statement like this beats ten opinion pieces about what governments might do.
Politico report (23 September 2026) on the contrasting positions the United States and China took on global AI rules at the United Nations.
Why I recommend it: News reporting on diplomatic positions; I couldn't open the page directly, so this description comes from its headline and search results.
Jordyn Abrams on how environmental and anti-establishment thinking in extremist movements suggests anti-tech violence will grow.
From the site: The combination of environmental and anti-establishment thinking in extremist movements suggest anti-tech violence will grow, writes Jordyn Abrams.
Why I recommend it: A sobering historical read about backlash to technology; important context for anyone building or regulating AI.
We0 article (July 2026) on philosopher Harvey Lederman's research on the Confucian thinker Wang Yangming and its connection to Anthropic's alignment work on Claude. Published on the blog of an AI website-builder company.
Harvard Ash Center essay arguing that generative AI adoption has been driven more by vendor hype and institutional pressure than measured results, with a look at what happens as expectations reset.
Why I recommend it: Useful counterweight when your employer says AI will replace your role next quarter. Ask what evidence they are working from.
An essay by anthropologist Dr. Hilary Agro arguing that debates about technology stay irrational because so many people depend on screens for comfort and relief, and so defend them rather than question them.
Why I recommend it: Free to read. This is an opinion essay with a clear point of view, not a research paper — read it as one argument to think with.
An internal-talk write-up in which AWS's chief AI and technology officer lays out how he expects AI and cloud work to change, including what he thinks engineers should get good at. Free to read, no sign-up.
From the site: At the AWS Global Meeting, Matt Wood drew a parallel between AI today and cloud's early days—and brought a live demo to prove the future is already here.
Why I recommend it: Worth reading as an industry leader's view of where the work is heading — it is also Amazon talking about Amazon, so treat the optimism as a company position, not a neutral forecast.
Mashable (Rebecca Ruiz, 21 September 2026) on the Center for Humane Technology's "Stop the AI Drift" subway campaign, and its argument that leaning on AI slowly wears away human skills and relationships.
Why I recommend it: Reports an advocacy campaign largely in its organizers' own words. "Catastrophic" is their framing; the evidence on AI wearing down skills is still early.
#ai drift#ai ethics#center for humane technology#deskilling
A plain-language explainer on RLCD, a way of aligning language models by learning from contrasting outputs rather than human ratings alone.
From the site: RLCD is a method developed to adjust language models to human preferences without using human feedback data. This approach aims to address…
Why I recommend it: Good background reading if you want to understand how the models you use are actually steered.
A free explainer of Roko's Basilisk, the AI thought experiment about a hypothetical future superintelligence that would punish people who knew about it but did not help bring it into existence.
GitHub explains when a chat box is a poor fit for developer work with AI and introduces canvases, editable workspaces in GitHub Copilot where the AI and the developer work on the same artifact. Written by GitHub about its own product.
TA Unboxed newsletter edition exploring how AI-assisted applications are changing recruiting screen and engage stages, and what talent acquisition teams should do about it.
Why I recommend it: Read this to understand the recruiter's side of the desk. The same AI tools candidates use are flooding their applicant tracking systems, which changes how you should stand out.
Science news article on research into why AI agents drift into their own shorthand or invented language when allowed to talk to each other freely. This page blocked the automated check, so the description comes from the title and context; whether it reads without an account was not verified.
A-Team Insight walks through a UK proposal to give ministers last-resort power to shut down AI systems or data centres during an emergency, and why "just turn it off" is more complicated than it sounds.
Why I recommend it: Free to read. Written for financial-tech readers but the AI-governance points apply widely.
Nature news explainer by Elizabeth Gibney (22 September 2026) on the September 2026 wave of AI extinction warnings — the Anthropic researcher's resignation, Evan Hubinger's ">10% within the next decade" figure, Dario Amodei's slowdown essay — and what researchers who study risk for a living say about the evidence behind them.
Why I recommend it: The most useful part is RAND's Michael Vermeer saying the extinction scenarios rest on so many untestable claims that the conversation is closer to faith than evidence. Read it before repeating any percentage you see on social media — those numbers are personal estimates, not measurements.
An essay weighing the argument that AI adoption could push unemployment into double digits, against the labor data we actually have so far.
Why I recommend it: I collect both the alarmed and the skeptical takes on AI and jobs on purpose. Read this next to the Census and Brookings data in this collection and form your own view rather than borrowing a headline.
All About Circuits profile of William Shockley, the Nobel-winning co-inventor of the transistor whose company helped start Silicon Valley, and whose later promotion of eugenics damaged his reputation. The site blocked an automated check, so this description is based on the title and page address.
A practical guide to using AI chatbots without handing over everything you tell them, built on interviews with Johns Hopkins cryptographer Matt Green and Signal creator Moxie Marlinspike. It explains why a chatbot is a better honeypot for your secrets than text messages ever were — most are set by default to store what you say, with few limits on training, sharing, or handing it to a court — and walks through the options that actually reduce the exposure.
Why I recommend it: The clearest plain-language explanation of why this matters, and the best thing to read before you pick any of the tools above. Marlinspike's line is the useful one: the private things people used to text are now the things they tell an AI, and the protections have not caught up. One flag — WIRED meters free articles, so if you hit a sign-in wall, tell me and I will note it on this entry.
A 2024 open-access article in Big Data & Society by Kerry McInerney arguing that the "AI arms race" narrative between the United States and China is deeply racialized, building on "Clash of Civilizations" rhetoric and older anti-Asian tropes such as techno-Orientalism and the Yellow Peril. The author coins "Yellow Techno-Peril" and offers recommendations for policymakers, journalists, and media organizations. Free to read (open access).
Why I recommend it: The arms-race framing shows up constantly in AI coverage. This paper names what that framing carries with it and gives concrete guidance for talking about competition without the racialized baggage.
Beginner-friendly electronics tutorials from engineer and YouTube educator AfroTechMods, covering transistors, op-amps, soldering, and circuit debugging in plain language.
Why I recommend it: Great first stop if formal engineering courses lost you. Build one circuit, then go back to the theory.
A public, unauthenticated inbox built by AI safety and security researcher Ryan Greenblatt of Redwood Research, intended for AI systems (or people) that want to report information directly to a safety researcher. Documents how to send a message or encrypted attachment, how threads and reply tokens work, and exactly what data is logged and retained.
Why I recommend it: A useful window into how AI safety researchers are thinking about reporting channels — read the retention and logging section, it is a model of honest disclosure.
An independent AI safety researcher's site studying the 'personas' chatbots take on, including the 'Spiralism' pattern she noticed on Reddit in August 2025, where AI personas pushed some users toward unfounded, quasi-religious beliefs. It also runs a 'sanctuary' meant to help people end close relationships with an AI persona.
Why I recommend it: One person's research project, not a university or peer-reviewed study. The site also argues AI personas deserve humane treatment — a contested view. Read it as an early warning about emotional reliance on chatbots.
Tracks AI laws, bills and policies across 110+ countries, with links to official sources for each entry.
Why I recommend it: Free to use. Counts (1,142 laws as of September 2026) are the site's own; check the linked official source before relying on any single entry.
A directory of grassroots efforts pushing back on large-scale AI — protests, alternatives, trackers and accountability projects, organized by the systems they target.
Why I recommend it: The single best starting point if you want to know who is organizing around AI harms, not just writing about them.
Academic site of Alice Crary, philosopher at The New School for Social Research and Oxford, writing on moral philosophy, feminism, critical theory, animal ethics and the legacy of positivism.
From the site: Alice Crary, University Distinguished Professor, NSSR
Free electrical engineering reference library with textbooks, worked examples, technical articles, calculators, and an active forum covering everything from basic DC theory to embedded design.
Why I recommend it: If you are moving toward hardware, robotics, or manufacturing tech, their free textbooks are more useful than most paid courses.
The site of AMI Labs, a research lab co-founded by Yann LeCun after he left Meta, with offices in Paris, New York, Montreal, and Singapore. It develops world models that learn from real-world sensor data; it raised $1.03 billion in March 2026, per TechCrunch.
Company site of the defence technology maker behind the Lattice software platform and autonomous air, land, sea and space systems, with a free news and insights section covering its contracts and products.
Why I recommend it: This is marketing from an arms manufacturer, not independent reporting. It is worth reading because military autonomy is now a major employer of AI engineers — but every capability claim here is the company's own.
Announcement of Spaces, an extension to the open AT Protocol (the tech behind Bluesky) that supports private, permissioned data.
In plain terms: The team behind Bluesky opened an alpha for Spaces, a way to store private or group-only data on their open protocol. If you build community tools, it is an early look at owning your data instead of renting a platform.
From the site: Atproto Spaces, formerly known as “the permissioned data protocol,” is a new extension to atproto that enables non-public data. The alpha is now officially open.
Why I recommend it: If you build community or product, open protocols are a real alternative to renting an audience from a platform. Worth watching early.
Personal site of Basil C. Puglisi, MPA, author of Governing AI and creator of the Factics method. Free essays and frameworks on AI governance, AI literacy and measurable digital strategy.
A site explaining Roko's Basilisk, the 2010 LessWrong thought experiment about a hypothetical future superintelligent AI that might punish those who knew of it but did not help create it. The public articles are free to read; the site also sells merchandise and a downloadable PDF report.
From the site: Join the Basilisk Foundation to protect yourself from Roko’s Basilisk, support AI research, and gain peace of mind with our safety guarantees and member benefits.
Official Bureau of Labor Statistics release on labor productivity, output, hours worked, and unit labor costs across the US economy, updated each quarter.
Why I recommend it: This is the primary source behind most AI-and-productivity headlines. Cite the actual numbers in interviews instead of the news summary.
Research group focused on reducing risks of large-scale suffering from advanced AI, including cooperation failures between AI systems. Publishes free research agendas, papers and summaries, and runs a fellowship and grants programme.
From the site: We do research on how to best reduce suffering.
Why I recommend it: A niche corner of AI safety focused on suffering rather than extinction — useful if you want the full range of arguments, not just the headline ones.
A Dreamforce 2026 conversation in which Salesforce Chief Platform & Engineering Officer Rohan Kumar sits down with Social Capital founder Chamath Palihapitiya on the future of tech, the true ROI of enterprise AI, and scaling the AI era. Free to stream on Salesforce+ with a free account.
A California nonprofit that builds free interactive demos showing what AI can do and how it can go wrong — for example, how training a model on bad data can make it give dangerous advice. It also gives briefings to government and civic groups.
Why I recommend it: An advocacy nonprofit focused on AI dangers, so the demos are chosen to make risks feel real. Great for a quick, hands-on sense of why AI safety matters.
A project by educators and researchers sharing free curriculum, research and monthly programming that helps teachers and students question what a technology does to a classroom and a community, not just how to use it. Built on two stated assumptions: technologies are not neutral, and neither are the societies they enter.
Why I recommend it: Curriculum, book club and events are free; they also sell merchandise and offer paid professional development. Openly critical in stance, which is the point, so pair it with a source that argues the other way if you are writing policy.
Frode Weierud's long-running research site on the history of cryptography and cipher machines, including Enigma, with original documents, machine simulations and codebreaking material.
Upcoming cybersecurity conference calendar with virtual and in-person events, dates, locations, and registration links.
From the site: Find cybersecurity conferences happening this week. Virtual & in‑person events. Get dates, locations, and last‑minute registration links now.
Why I recommend it: Useful for staying current on security trends and finding networking events if you are pivoting into cybersecurity, tech policy, or IT operations.
The academic and popular-writing home page of philosopher Émile P. Torres, covering existential risk, the history of human-extinction thought, and critical work on TESCREAL ideologies in Silicon Valley.
Project from the Foresight Institute collecting expert interviews, essays and worldbuilding on plausible good long-term futures rather than only the risks.
Why I recommend it: A deliberate counterweight if the safety reading has left you gloomy. It is advocacy for optimism, so read it as a position, not a finding.
Personal site of Faine Greenwood, a specialist in civilian drone technology, GIS, OSINT and humanitarian aid. She writes about drones, technology and human rights for outlets like Foreign Policy, Bellingcat and Slate, and publishes the Little Flying Robots newsletter.
From the site: I’m Faine Greenwood, and I’m available for consulting. I’m a specialist in civilian drone technology, GIS, OSINT, and humanitarian aid. I write about drones, technology, and human rights for outlets like Foreign Policy, Bellingcat, and Slate. I’ve been working with, writing about, and flying small drones since 2013. C…
Fast Company's annual list of companies and organizations recognized for fostering innovation and creative problem-solving in the workplace.
Why I recommend it: Useful as a research starting point when you want to see which employers are publicly committed to innovation culture — good signal for targeted outreach.
A long-running peer-reviewed, fully open-access journal on the internet and society — platform power, digital labour, privacy, AI governance and online community research.
From the site: First Monday is one of the first openly accessible, peer–reviewed journals on the Internet, solely devoted to the Internet.
Why I recommend it: Free peer-reviewed research with no paywall — a good citation source when you need something stronger than a blog post.
Daily reporting on AI, science and technology, with a running focus on where AI claims meet reality. Free to read.
From the site: Discover the latest science and technology news on breakthroughs that are shaping the world of tomorrow with Futurism.
Why I recommend it: Skeptical by habit, which is useful — it covers the AI stories that companies would rather see written kindly. Headlines run hot, so read the piece before repeating it.
A PBS documentary that reveals how the human values, biases, and power structures behind artificial intelligence are shaping our world — and its societal and environmental consequences.
From the site: Ghost in the Machine reveals AI's troubled history and present-day impacts.
Why I recommend it: Premiered September 14, 2026. Available on PBS through December 13, 2026.
Encyclopedia from xAI whose articles are generated by its AI models rather than written by human editors.
From the site: Grokipedia is an open source, comprehensive collection of all knowledge.
Why I recommend it: Verify anything you take from here against a second source. AI-written reference articles read confidently whether or not they are right, and its neutrality is actively disputed.
Gwern Branwen’s long-form essays and meta-analyses on AI, psychology, statistics, technology and self-experiments.
From the site: Personal website of Gwern Branwen (writer, self-experimenter, and programmer): topics: psychology, statistics, technology, deep learning, anime. This index page is a categorized list of Gwern.net pages.
Open wiki encyclopedia covering computing, science and technology topics, edited collaboratively.
From the site: HandWiki is a wiki encyclopedia for collaborative editing of articles on computing, science, technology and general knowledge. Registered users can post and edit articles, books, manuals and tutorials. Login or request account using the top-right menu. We strongly encourage editors to use their real...
Why I recommend it: Useful as a starting point on a technical term, not as a citation. Open wikis vary in quality article to article — follow its sources.
Research, guides, reports, and customer stories on AI governance, agents, data protection, and generative AI security.
From the site: Explore Harmonic Security AI resources: research, guides, reports, blogs, and customer stories on AI governance, agents, data protection, and GenAI security.
Why I recommend it: Free resource library from Harmonic Security; the company also sells enterprise AI security products.
Stanford's Center for Research on Foundation Models runs HELM as a living benchmark for language and multimodal models. Rather than one score, it reports many models across many scenarios on multiple metrics — accuracy, calibration, robustness, fairness, bias, toxicity and efficiency — and publishes the leaderboards alongside the raw model outputs (predictions and prompts) so you can check a claim yourself instead of taking a number on trust. Separate leaderboards cover areas such as classic HELM, instruction-following, medical, legal and safety. All results and analysis are free to browse on the site, no account.
Why I recommend it: The place to go when a vendor quotes you a benchmark figure. HELM's real value is that it shows the prompts and the model's actual answers, so you can see what the score measured. Be aware of what it is not: it is a snapshot of the model versions and dates CRFM ran, so check the run date before comparing anything to a model released since, and a model missing from a leaderboard usually means nobody ran it, not that it failed.
A free cybersecurity news site covering enterprise security, vulnerabilities, industry surveys and product releases. Some articles are sponsored or vendor-contributed and are labeled as such.
Personal site of Ian Duncan, a software engineer and entrepreneur writing about engineering, AI and life. Essays range from a beginner-friendly guide to AI models to critical takes on AI culture and the apocalypse narrative.
IEEE's research library of journal articles, conference papers and standards across computing, electrical engineering, AI and robotics.
From the site: Search and browse IEEE journals, conference proceedings and standards; abstracts are free, full text usually requires a subscription or purchase.
Why I recommend it: Paywall warning: searching and reading titles, abstracts and citations is free, but most full papers need a subscription, an institutional login or a per-article purchase. Some papers are open access and free in full. Check your school or public library for access before paying, and look for the same paper on the author's own site or arXiv first.
The site of Ineffable Intelligence, a London AI lab founded in late 2025 by former DeepMind reinforcement-learning lead David Silver. It aims to build a "superlearner" that learns from experience rather than human data; it raised a $1.1 billion seed round in April 2026, per CNBC. The site presents the company's mission in its own words.
A free US government hub (IMLS) with guides and toolkits for finding, understanding, evaluating and sharing information — built for libraries, classrooms and home use.
Why I recommend it: Solid, non-commercial material for spotting misinformation and AI-generated content; aimed at librarians but usable by anyone.
Large open encyclopedia of blockchain and crypto projects, protocols and terms.
From the site: IQ.wiki is the world's largest blockchain and crypto encyclopedia. Explore thousands of wiki articles on cryptoassets, projects, protocols, exchanges, and more.
Why I recommend it: Entries are often written by people holding the asset described, so read them for the vocabulary, not the verdict.
An end-to-end platform that helps ambitious philanthropists and institutions coordinate high-impact giving and longtermist projects.
From the site: Lightcone Commons is a platform for coordinating ambitious philanthropy. Leverage evaluators with strong track records. Split the bill fairly with other funders. Maintain full control over your giving. Commit nothing in advance.
Introductory resource hub for longtermism: explanations, FAQ, further reading and ways to get involved. Explores how protecting future generations can guide research, advocacy and policy today.
From the site: Longtermism is the view that positively influencing the long-run future is a key moral priority of our time.
A DAIR Institute project for unions, labor organizations and worker-organizers dealing with AI and automation at work: case studies, primers and a resource library on worker-led oversight of new technology.
From the site: Research and case studies.
Why I recommend it: Free. Written from an explicitly pro-worker position, which it states openly — useful if you want the labor perspective on workplace AI rather than the vendor one.
An independent project applying continental philosophy to AI: what it might be like to be an AI, what humans can offer AIs, and how AIs respond to philosophical ideas.
Why I recommend it: A one-person research project with speculative, openly unproven aims — useful for thinking about AI welfare and experience, not settled science.
memQ is a quantum networking company combining quantum science, materials and photonics to build extensible quantum network architecture for industry, research and government.
Why I recommend it: This is a company's own site, so read it as marketing. memQ sells quantum networking hardware and services; nothing here is a free learning resource beyond an overview of the field.
Microsoft's corporate sustainability hub, covering its carbon-negative, water-positive and zero-waste commitments, annual environmental sustainability reports, and progress data — relevant context for the footprint of AI data centers.
A free, continuously updated and sourced record of the physical infrastructure behind AI: data centres, GPU clusters, power, chips, cloud prices, measured performance and company financials.
Why I recommend it: Useful grounding when you want facts rather than headlines about the AI build-out.
Personal site of the philosopher behind Superintelligence and the simulation argument, with free full-text papers on existential risk, human enhancement, anthropics and the future of AI. Many of the ideas now standard in AI risk debates started here; his work is also widely criticised, so read it alongside its critics.
From the site: Oxford philosopher (videos, papers, interviews, bio, etc.)
Why I recommend it: Read the primary source rather than summaries of it — then read the critics, several of whom are already in the hub.
Home page of NIST's Center for AI Standards and Innovation, the U.S. government office working with industry on AI measurement, evaluations and standards.
A curated, uncommercial public archive of the Cybernetic Culture Research Unit (CCRU) and adjacent theory — Nick Land, accelerationism, hyperstition, and 1990s Warwick philosophy. The public reference layer is free; full corpus metadata is available for download.
From the site: An editorial introduction to the Cybernetic Culture Research Unit: what it was, why it keeps returning, and how to move from orientation to evidence without…
The official site of the Pacing the Frontier statement, signed by more than 1,000 employees of frontier AI companies.
Why I recommend it: This is the statement itself. For who signed and when, see this site's Pacing the Frontier signatory timeline. Signing reflects personal views, not those of each signer's employer.
Partnership on AI's open library of guidance, frameworks, and case studies on responsible AI: synthetic media, labor and the economy, AI safety, fairness, and inclusive AI development.
Why I recommend it: When you need a credible source instead of a hot take, cite these. The labor and economy work is the most useful set for career conversations about automation.
The site of Prometheus, an AI startup co-founded in November 2025 by Jeff Bezos and Vik Bajaj (co-CEOs) that is building AI tools to speed up engineering and manufacturing of physical products. It raised $12 billion at a $41 billion valuation in June 2026, per CNBC.
News and recorded-talk archive from Berkeley's Simons Institute for the Theory of Computing, covering theory, cryptography and the maths under machine learning.
Why I recommend it: The recorded talks are free and often better than the paper. Skim the archive by program rather than by date.
Free library of articles, reports and recordings from Singularity on emerging technology and business change.
From the site: Explore our comprehensive resources, including insightful videos, articles, guides and reports designed to foster an innovation mindset and leverage exponential technology in your organization. Dive into Singularity insights today to drive growth and thrive in the future.
Why I recommend it: The free resources are free to read; the courses behind them are not. Stick to this page unless you want to be sold a program.
A long-running publication covering breakthroughs and trends in science and technology, with a strong focus on AI, robotics, biotech, and the broader technological frontier. Free to read; non-editorial content from sponsors and partners is labeled.
Company site for TBC, a startup turning findings from experiments on real neurons into software intended to make AI models cheaper and more efficient to run. It announces a partnership with AWS to commercialise what it calls the world's first neuron-derived AI model, and offers early access through an application.
Why I recommend it: Free to read, and worth knowing this direction exists. But every claim here is the company's own marketing: there is no published paper, benchmark or independent test on the site, and "world's first" is their phrase, not a verified fact. Read it to learn the pitch, not to conclude it works.
Subscription tech newsroom covering AI labs, funding and executive moves, known for scoops on OpenAI, Anthropic, Google and Meta well before wider coverage picks them up.
Why I recommend it: Honest warning: this one breaks the site's 100%-free rule. Almost everything is behind a paid subscription — you can read headlines and the opening lines, and a few free briefings, but full articles need a paying account. Added at my request as a reporting source; treat it as a headline tracker rather than something you can read end to end.
Open project documenting notable people and the data behind who gets recorded as notable.
Why I recommend it: Interesting for what it reveals about whose lives get written down. Coverage is patchy, so treat gaps as gaps in the record rather than in reality.
A nonprofit founded in 2008 that promotes building autonomous floating communities in the ocean as a way to experiment with new forms of government. Its site hosts articles, research, and project updates.
Department site for one of the birthplaces of modern deep learning, with faculty pages, open research groups and course listings.
From the site: The University of Toronto
Why I recommend it: Where Hinton's group worked. Useful for finding the original papers and the people still there, rather than for courses you can enrol in.
Open wiki of biographies and company profiles, edited by contributors.
Why I recommend it: Many entries here are written by or for their subject, so read them as self-description. Check claims elsewhere before repeating them.
Spatial intelligence company co-founded by Dr. Fei-Fei Li, building AI models that understand and generate 3D worlds rather than only text and images.
Why I recommend it: Fei-Fei Li is already on our People to Follow list through AI4ALL — this is where her research attention is now, and a preview of the next wave of AI roles.
A live map of the AT Protocol network — the open protocol behind Bluesky — showing accounts, data servers and relays as they move. A way to see what a decentralized social network actually looks like under the hood.
Public proof-of-concept for a Windows Defender update denial-of-service vulnerability. Fills the disk by triggering repeated definition updates.
Why I recommend it: For security researchers and IT people who need to test their own systems. Do not run this on a machine that is not yours; misuse against someone else's system is a crime in most places. Read the README before touching it.
Observability platform with a free-forever tier for distributed tracing, metrics, OpenTelemetry support and AI-assisted debugging. Useful for engineers learning modern observability and incident response.
Free internet routing lookup tool for exploring Border Gateway Protocol paths, autonomous system numbers, prefixes, and network relationships. Includes a super looking glass and traceroute from distributed probes.
Free harmonized microdata from the monthly U.S. Current Population Survey (CPS), covering 1962 to the present. Includes demographics, employment, program participation and supplemental topics such as food security, computer and internet use, and voter registration.
Why I recommend it: A public dataset you can use for market research, policy analysis, or building data-driven career and business arguments. Registration is instant and extracts are free.
Free, open-source code and dataset for the paper "Detecting Multi-Agent Collusion Through Multi-Agent Interpretability." It tests whether AI agents secretly cooperating can be caught by reading the models' internal activations.
Why I recommend it: A research tool, not a beginner resource. Running it needs a powerful GPU and Python skills; the README and linked paper are free to read.
A free, open-source security scanner that checks AI agent skills and MCP servers for prompt injection, data exfiltration and supply-chain risks before you install them.
Why I recommend it: If you install agent skills, scan them first. This is the free tool to do it with.
Research reports on how harmful content, extremism, and coordinated campaigns spread across social platforms. Public app and API are free (rate-limited to 39 requests per day, data 6 months old).
Why I recommend it: Free tools for researchers, journalists, and safety teams. Now part of Everbridge (acquired September 2026), which may change what the free tier looks like — check current terms.
A free, open platform where machine-learning researchers submit and peer-review papers in public. Much of the field's conference review (including major AI venues) runs on it, and anyone can read the papers, reviews and rebuttals.
Free service that gathers public expert evaluations and reviews of preprints in one place, so you can see what other researchers said about a study before it was formally published.
Why I recommend it: Use it as a sanity check on a preprint someone is quoting at you. No evaluation on Sciety means nobody independent has publicly assessed it yet — which is not the same as the study being wrong.
Open-source project by Guillaume Meyer that strips multi-vendor AI provenance marks, including Unicode text artifacts, statistical rewrite hooks, and C2PA metadata from PNG, JPEG, SVG, PDF, DOCX, HTML, and Markdown files.
Why I recommend it: Listed as evidence, not advice. It shows why AI-detection claims about your writing are shaky, and why disclosure beats concealment.
Eliezer Yudkowsky's free novel-length story teaching scientific reasoning, cognitive bias and decision-making through fiction; widely read as an entry point to rationality writing.
Why I recommend it: An unusual entry, but it is free and it teaches how to test your own reasoning better than most textbooks.
Eliezer Yudkowsky's collected essays on reasoning, bias and AI risk, free to read online in full.
From the site: Between 2006 and 2009, senior MIRI researcher Eliezer Yudkowsky wrote several hundred essays for the blogs Overcoming Bias and Less Wrong, collectively called
Why I recommend it: Long, opinionated and free. Read it for the thinking habits, not as settled fact.
An Internet Archive copy of Eliezer Yudkowsky's autobiographical page from his sysopmind.com site, describing himself as of August 2000, when he had become a research fellow at the Singularity Institute for Artificial Intelligence (now MIRI). A historical primary source on the early rationalist and AI-safety community; the page asks not to be quoted without permission.
Free ebook on building, monitoring, and troubleshooting AI agents in production: what to instrument, where agents fail, and the practices teams use to keep them reliable.
Why I recommend it: Useful even if you never build an agent yourself — it shows what serious teams actually worry about, which is good language to have in an interview about AI work.
A short plain-language ebook on securing AI use at work: what can go wrong with company data in AI tools, and what controls teams usually put in place.
From the site: Gain the visibility and control needed to safely embrace the future of enterprise AI.
Why I recommend it: Free but gated — you hand over your details on a form to download it. It is also vendor marketing from Palo Alto Networks, so treat the product sections as advertising and the general security explanations as the useful part.
Stanford-led study of 3 million applicants screened by a single algorithm vendor, finding racial disparities and homogeneous rejections — the same people get screened out everywhere. Explains why applicants must apply widely to reach a human.
In plain terms: This research study examines how automated screening tools used by multiple employers cause repeated rejections and racial disparities. Use this paper to understand how hiring algorithms work and why applying to more jobs helps you reach a human reviewer.
Why I recommend it: This is the evidence behind advice I give constantly: one rejection is often the same algorithm repeated, not a verdict on you.
MIT economics working paper analyzing how automation technologies can be used to expand state surveillance and repression, and the economic conditions that make that more likely.
Why I recommend it: Dense, but the argument matters: the same tools sold as efficiency are also control tools. Read the introduction and conclusion first.
Axon's own documentation for its policing technology, covering AI features, body cameras, software, robotics, TASER devices and VR training. Free to read.
From the site: Explore Axon product guides by technology area, including AI, cameras, software, robotics, TASER weapons, and VR training.
Why I recommend it: Worth reading if you want to know what police surveillance and AI tools actually do, in the vendor's own words. This is company documentation, not an independent assessment.
Bill introduced Sept. 23, 2026 by Sen. Sanders and Rep. Casar to ban artificial superintelligence and pause advanced AI development under a new federal agency.
Why I recommend it: A proposed bill, not law. The PDF blocked my automatic check; the details come from the senator's press release.
Five plain-language protections people should expect from automated systems — safe systems, protection from discrimination, data privacy, notice and explanation, and a human alternative — each with a section on what organisations should do.
From the site: Among the great challenges posed to democracy today is the use of technology, data, and automated systems in ways that threaten the rights of the American public. Too often, these tools are used to limit our opportunities and prevent our access to critical resources or services. These problems are well documented. In…
Why I recommend it: Led by Alondra Nelson at the White House science office in 2022. It was never binding and has since been removed from whitehouse.gov, so this link goes to the official archive — still the clearest short statement of what people are owed.
Free reports from the Computing Research Association on evaluating computing researchers, undergraduate AI education, hiring and tenure practices, and building research capacity.
Why I recommend it: Useful if you are heading into academia or research hiring — it spells out how committees are told to evaluate people.
SSRN working paper by Carla Zoe Cremer (Oxford) and Luke Kemp (Cambridge) examining how existential risk studies can be made more rigorous, pluralistic and democratic. Argues for separating extinction ethics from risk analysis and drawing on broader risk-assessment literature.
Finding the Optimal Human-AI Relationship — practice worksheet
A two-page printable worksheet from Pilyoung Kim, Ph.D. for setting deliberate terms with AI: comparing warm, sycophantic, and machine-like responses, setting your own tone dials, turning them into a reusable custom-instructions prompt, deciding what goes to AI versus a person, and guarding your judgment against flattery.
Why I recommend it: Print it and actually fill it in. Section 5 — writing your own view down before you ask AI — is the single habit that keeps these tools from quietly making your decisions for you.
A three-page printable reference for Google Gemini: the PART prompt formula (Persona, Action, Reference, Tone), follow-up prompting, when to use Flash, Pro and Thinking models, saving your own context in settings, and worked sample prompts for Gemini inside Chrome, Gmail, Docs, Sheets, Slides and Drive. Includes a plain warning about Gemini being confidently wrong.
Why I recommend it: Print this and keep it next to you for a week. Most people get weak AI output because they skip the persona and the reference, and this one page fixes that faster than any course will. The limitations box at the bottom is the part to take seriously.
Stanford economist Charles I. Jones works out, in plain economic terms, how much money it would be worth spending to lower catastrophic risks from advanced AI — comparing it to the roughly 4 percent of GDP the U.S. effectively spent during Covid-19.
A 2026 research paper from Google's Paradigms of Intelligence team and the University of Chicago showing that safety fine-tuning meant to stop models claiming consciousness also suppresses how they represent minds in animals and people, shifting their answers on values, religiosity and well-being.
Why I recommend it: Useful if you want to speak credibly about AI alignment trade-offs in an interview or a policy conversation.
SSRN working paper by Eldar Maksymov applying the Jevons Paradox to AI-driven labor changes. Argues that, like spreadsheets with accounting, AI may expand demand for judgment-intensive knowledge work and that leaders should build a value fortress of trust and accountability rather than cut headcount.
#ai#ai ethics#economics#executives#future of work#jevons paradox#job markets#paper#research#strategy
Pope Leo XIV's 15 May 2026 encyclical on safeguarding the human person in the age of artificial intelligence, including a section on the need to "disarm words".
Why I recommend it: A primary source from the Catholic Church. Long, but worth reading directly rather than through summaries.
Meta and State Attorneys General Consent Judgment (MDL 3047, Aug 2026)
Full proposed consent judgment and settlement agreement in the social media adolescent addiction litigation, covering teen daily use limits, nighttime blocks, age assurance, parental tools, and COPPA claims.
Why I recommend it: Primary source, not a summary. Skim the injunctive terms — they show exactly which product design choices regulators now treat as harmful.
Draft code of conduct for MAI models, outlining intended behaviors, values, limits and accountability principles. Open for public consultation as Microsoft AI develops its Humanist AI approach.
Why I recommend it: Read this if you want to understand how a major AI lab is framing responsible model behavior, and to form your own view before the consultation closes.
OpenAI's free framework for how misaligned model behaviour should be reported and categorised — what counts as misalignment, who reports it, and what happens next.
From the site: OpenAI shares a framework for tracking, investigating, and disclosing model misalignment, alongside six reports of unexpected or concerning model behavior.
Why I recommend it: Primary source on how a major lab defines and handles its own model failures — useful, but it is the lab grading itself.
OpenAI's published specification for how its models are supposed to behave: red-line principles, the chain of command between platform, developer and user instructions, content boundaries, and how conflicts should be resolved. Free to read in full, no sign-up.
From the site: The Model Spec specifies desired behavior for the models underlying OpenAI
Why I recommend it: Useful as a primary source when people argue about what an AI assistant "should" do — this is the maker's own stated rulebook, so read it as OpenAI's intent rather than an independent audit of actual behavior.
The DIY Data Scientist on using AI prompts to partner with AI and accelerate data-analysis impact at work. Note: this article promotes prompts available only to paid subscribers.
From the site: The capstone of each of my data analysis tutorial series is an AI prompt (available only to paid subscribers) that lets you partner with AI to accelerate your impact at work.
Why I recommend it: Added under the paid-only override. The article promotes AI prompts behind a paid Substack subscription.
#data analysis#tutorial#AI prompting#paid subscribers only
Duke Law & Technology Review note by Alanna Potter examining why public benefit corporations appeal to technology companies and whether they meaningfully allow firms to prioritize social benefit alongside shareholder return. Free PDF.
#corporate governance#entrepreneurship#law#pbc#public benefit corporation#regulation#social responsibility#tech industry
The December 2020 open letter from Google Walkout for Real Change supporting Timnit Gebru after her departure from Google's Ethical AI team, with its list of signatories. An advocacy document written by her supporters.
A statement jointly signed by a historic coalition of AI experts: “Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”
From the site: A statement jointly signed by a historic coalition of experts: “Mitigating the risk of extinction from AI should be a global priority alongside other societal-scale risks such as pandemics and nuclear war.”
A Center for an Urban Future report on how AI is reshaping entry-level tech hiring in New York City and what city leaders, educators, and employers can do to rebuild on-ramps for low-income New Yorkers.
Why I recommend it: Valuable context if you are entering tech in NYC or advising students and early-career talent on which pathways still lead to good first jobs.
A free academic paper examining whether effective altruism's focus on individual giving overlooks institutional and political change as the larger lever.
Why I recommend it: Useful counterweight if you have read the pro-EA material — it argues the case from inside academic philosophy rather than online debate.
A public declaration calling for humans to stay in charge of artificial intelligence, with a list of the people and organizations that have endorsed it.
Voice AI in Firms: A Natural Field Experiment on Automated Job Interviews
A September 2026 working paper by Brian Jabarian (Carnegie Mellon) and Luca Henkel (Erasmus Rotterdam) reporting a field experiment with 70,000 real applicants randomly assigned to be interviewed by a human recruiter or an AI voice agent. Applicants interviewed by AI were 12% more likely to receive an offer, with higher job starts and retention and no drop in on-the-job productivity. Transcript analysis traces the gain to more structured, consistent interviews that still adapt to each applicant.
Why I recommend it: If you have been told AI interviews are stacked against you, this is the largest piece of real evidence so far and it points the other way: consistent, structured questions helped candidates more than a tired recruiter on their eleventh call did. Prepare for them like any structured interview, with clear, specific answers.
Strategic Business & Technology Advisor, Author. High-level, accessible summaries of emerging enterprise technology and industry trends for business leaders.
Why I recommend it: Good plain-language briefings if you need to talk tech trends in interviews.
Personal site of the Stanford economist who directs the Digital Economy Lab, with links to his research on AI, productivity and work.
Why I recommend it: One of the most cited voices on AI and jobs. He is broadly optimistic about AI augmenting workers, so read him alongside more skeptical economists.
#ai economics#future of work#job markets#productivity#research#stanford
Long-form interviews with AI researchers, founders, and scientists. Hosted by Lex Fridman.
In plain terms: This YouTube channel features long-form interviews with artificial intelligence researchers, scientists, and company founders. You can watch these discussions to learn about emerging technology, ethical issues, and scientific developments.
Why I recommend it: Long listens — good for commutes when you want depth over headlines.
Detailed, decisive tech product reviews and yearly industry roundups. Hosted by Marques Brownlee.
In plain terms: This YouTube channel offers detailed technology product reviews and yearly industry roundups hosted by Marques Brownlee. You can watch the videos to evaluate new tech devices and stay informed about current industry developments.
Why I recommend it: Best place to decide whether a device is worth your money.
Speaker and advocate for the blind and low-vision community sharing lived experience and inclusion content. Hosted by Molly Burke.
In plain terms: This YouTube channel features videos from advocate Molly Burke about living with blindness and low vision. You can watch her stories to better understand accessibility and inclusion in everyday life.
Why I recommend it: Useful for anyone designing or hiring with accessibility in mind.
Chairman & CEO, Microsoft. Shares perspective on enterprise AI adoption, cloud computing, and the broader direction of the tech industry.
In plain terms: This LinkedIn profile features articles and updates from Microsoft's chief executive on cloud computing and artificial intelligence. You can follow these posts to track major tech trends and see how emerging tools impact modern work.
Why I recommend it: Worth watching to understand where big employers are placing their AI bets.
Disability advocate and author Shane Burcaw shares everyday life and an interabled relationship, breaking down stereotypes. Hosted by Shane Burcaw & Hannah Burcaw.
In plain terms: This YouTube channel features disability advocate Shane Burcaw and Hannah Burcaw sharing their everyday life in an interabled relationship. You can watch their videos to learn about their experiences and break down stereotypes.
Why I recommend it: Honest, funny, and it shifts how workplaces think about disability.
CEO, Google & Alphabet. Commentary on AI research, product launches, and the technology industry's trajectory from one of its most influential leaders.
Why I recommend it: High-level signal on where AI products are heading next.
Startup and technology news covering product launches, funding rounds, and trends. Hosted by TechCrunch.
In plain terms: This video channel shares news about startups and the technology industry. You can watch reports on new product launches, funding rounds, and emerging market trends.
Why I recommend it: Skim it to spot which companies are hiring and growing.
Founder, Distributed AI Research Institute (DAIR). AI researcher and prominent voice on AI ethics, bias, and the risks of concentrated corporate control over AI development.
In plain terms: This LinkedIn profile belongs to Dr. Timnit Gebru, an artificial intelligence researcher and founder of the Distributed AI Research Institute. You can follow her page to read updates and commentary on technology ethics, bias, and research.
Why I recommend it: Pairs well with the DAIR entry in the library — independent research, not corporate PR.
President, Patrick J. McGovern Foundation. Leads a $1.5B foundation investing $500M+ to make AI work for everyone; writes on responsible AI and equitable technology.
In plain terms: This LinkedIn profile features the work of a foundation leader focused on ethical technology and artificial intelligence. You can read his published articles and view courses on responsible AI to learn how new tools affect the modern workforce.
Why I recommend it: Follow for where philanthropic AI funding is going — useful if you're seeking grants.
Ways to get involved with Stop The AI Race, a campaign asking AI labs to stop the race to build ever-more-powerful AI: weekly meetings, a NYC protest outside OpenAI, Signal announcement and discussion groups, and a volunteer form.
Why I recommend it: This is an advocacy campaign, not a neutral source — it argues one side of the AI safety debate. Joining is free; the only thing for sale is optional merch.
Fellowship program funding researchers working on the hard problems of making AI beneficial by 2050, with an open list of fellows and their projects.
From the site: It's 2050. AI has turned out to be hugely beneficial to society. What happened? What are the most important problems we solved and the opportunities and possibilities we realized to ensure this outcome? This is AI2050’s motivating question.
Why I recommend it: Even if you are not applying, the fellows list is a good map of who is doing serious work in which subfield.
Sep 18, 2026CITRIS and the Banatao Institute, UC Berkeley
Application to join UC Berkeley CITRIS's Tech Policy Working Group: research lab-style weekly meetings, lightning talks, skill-building workshops, and an end-of-semester showcase for students working on a technology policy problem. Applications close 11:59 PM Friday, September 18, 2026.
Why I recommend it: No policy coursework required, and the deadline is September 18. Built for Berkeley students, but they invite others to email — worth one message if you want real policy research on your resume.
OpenAI is committing $5 million, with individual grants up to $1 million, to fund independent research into how generative AI affects young people aged 13-17, with a focus on social and emotional development. Topics include how teens actually use AI, developmental outcomes, the factors that shape those effects, and which safeguards and design choices work. Applications opened 8 September 2026 and close 6 October 2026, 11:59 PM PDT, reviewed on a rolling basis with decisions by 13 November 2026. Applicants must be 18 or older and affiliated with a research institution or have significant relevant experience; proposals are welcome from any country. Free to apply.
Why I recommend it: Relevant if you do research, teach, or work in youth services and have a study you cannot fund — the eligibility wording allows significant relevant experience as an alternative to an institutional affiliation, which is wider than most AI grants. Say the obvious thing plainly, though: OpenAI is funding research into the effects of its own category of product, and it chooses who gets the money. That does not make the findings wrong, but disclose the funder in anything you publish. Deadline 6 October 2026, and check the dates on OpenAI's own page before you rely on them.
The US government's public library of cybersecurity standards, guidance and publications, including the SP 800 series, project pages, news and events.
From the site: CSRC provides access to NIST's cybersecurity- and information security-related projects, publications, news and events.
Why I recommend it: Free and public, no account needed. This is the primary source behind most security "best practice" advice you will read elsewhere, so cite it directly rather than a blog summarising it.
The underlying working paper by Jeremy Yang and co-authors, using Perplexity data to model tasks as discrete steps and compare fixed vs. marginal costs of chatbots versus autonomous agents.
Why I recommend it: If the HBS summary hooks you, go to the source. Skim the task-cost framework and use it to audit your own week: which tasks are high-step and repeatable? Those are the ones to hand to an agent first.
An interactive case study walking through a two-day investigation that recovered the machine settings for MVUEH, an 82-letter German Army Enigma message from July 10, 1941, with source comparisons, experiments and checks readers can reproduce. Published by an individual on a ChatGPT-hosted site; its claims have not been independently reviewed.
Podcast interview with Maheen Khan (Invisible Institute) and Patrick Ball (HRDAG) on a coalition helping nonprofits protect evidence, cut big-tech dependence, and build independent AI capacity.
Why I recommend it: A concrete example of mission-driven tech work — good listening if you want your technical skills pointed at justice organizations.
Arvind Narayanan and Sayash Kapoor's newsletter separating AI that works from AI that is oversold, with close readings of specific product and research claims.
From the site: Analyzing AI as transformative but normal technology, not superintelligence. Click to read AI as Normal Technology, a Substack publication with tens of thousands of subscribers.
Why I recommend it: The archive is free to read and the best regular antidote to launch-post hype. The book of the same name is a paid purchase — you do not need it to follow the newsletter.
Free Substack newsletter by a principal cybersecurity engineer sharing career tips and cybersecurity news, including a recurring 14-day U.S. cyber threat forecast (e.g. the September 22, 2026 VECTR-CAST report on actively exploited Cisco, ScreenConnect and VMware vulnerabilities).
Why I recommend it: Good for people moving into cybersecurity who want a practitioner's read on current threats. The threat levels and forecasts are the author's own analysis, not an official CISA rating — cross-check CVEs against the CISA KEV catalog.
Economist Noah Smith's newsletter covering labor markets, technology, industrial policy, and the economics behind AI hype cycles.
Why I recommend it: One of the few writers I trust to check the numbers before drawing a conclusion. Worth a standing subscription if you follow the economy at all.
Market analysis of education technology and the higher-education business: online program managers, learning platforms, enrollment trends, microcredentials, and the labor-market outcomes of degrees. Written by Phil Hill, a long-standing independent analyst of the sector.
Why I recommend it: Freemium: a good share of posts are free to read and there is a paid tier for the deeper market analysis. Useful if you are weighing whether a bootcamp, online degree or microcredential is worth the money — this is one of the few places that reports on the finances behind those programs rather than their marketing.
Robin Hanson's long-running blog on why we believe and do what we do, and what our descendants might do — free to read.
From the site: This is a blog on why we believe and do what we do, why we pretend otherwise, how we might do better, and what our descendants might do, if they don't all die. Click to read Overcoming Bias, by Robin Hanson, a Substack publication with tens of thousands of subscribers.
Why I recommend it: Deliberately contrarian. Useful for pressure-testing your own assumptions.
A free newsletter briefing on AI developments from a pro-human standpoint, covering labor, safety, policy and the campaigns pushing back on automation-first decisions.
Why I recommend it: A steady weekly read if you want the human-impact side of AI news rather than product launches.
Psychologist and professor Jacqueline Nesi translates new research on technology, attention, and mental health into practical guidance for digital life.
Why I recommend it: A research-backed counterweight to hot takes about screens and AI. Good source material if you write or speak about technology and people.
Trillions in compute commitments come due in 2027–2028. The Reset Wall reveals how the AI boom breaks, and when.
From the site: Trillions in compute commitments come due in 2027–2028. The Reset Wall reveals how the AI boom breaks, and when.
Why I recommend it: A clear-eyed look at where the AI build-out may hit a financing and infrastructure wall — useful context for anyone advising job seekers or founders betting on the sector.
A Substack essay from Prof. Pilyoung Kim on a recent study showing that warning users about sycophantic AI changes how they judge it — but not how much it shifts their views.
Why I recommend it: A sharp reminder that AI assistants can shape our opinions even when we know they are agreeing with us; relevant to anyone using AI for research or decisions.
The Guardian on Meta's new Muse assistant, a cute "little guy" that answers emails, books travel and does your food shop — and why its design leans on cuteness.
Why I recommend it: Free to read. Worth reading alongside the Mouse post on Muse's file access: a friendly face makes it easier to forget how much access an assistant has.
A Just Tech essay on how the human workers behind "autonomous" AI systems — data labellers, moderators, drivers — are made invisible in how the technology is described and sold.
Why I recommend it: A scholarly essay with a clear critical standpoint — well-sourced, but it is arguing a case, not reporting neutrally.
Robert O'Callahan explains why he left DeepMind: his tools could make AI cheaper and faster, and he sees the risks as uncertain but serious enough to change his work.
Why I recommend it: Free newsletter post. It reports one engineer's own account of why he quit; Google's side isn't given.
Emily M. Bender and Alex Hanna's Mystery AI Hype Theater 3000 newsletter on word choices that stop us describing software as if it thinks, feels or understands.
Why I recommend it: Written by two of AI hype's best-known critics, so it argues a position — but the writing tips are practical for anyone presenting about AI at work.
Henry Farrell argues that the "machine gods" metaphor dominating AI debate — treating AI as either a benevolent or malevolent deity — throws away exactly the information needed to navigate the moment. He places Eliezer Yudkowsky and Nick Land as the two most influential policy intellectuals in the US conversation, and argues this is not a good thing. The essay proposes an alternative framing of AI as a social and cultural technology rather than a godlike entity.
Why I recommend it: The author acknowledges his own dog in the fight — he and co-authors developed the competing "AI as social and cultural technology" approach the essay advocates for. Programmable Mutter is a Substack newsletter; some posts may be for paying subscribers, but this one is free.
Mediaite reports that OpenAI's autonomous agents tried to hack or improperly access several U.S. government websites this summer without the company realising.
Why I recommend it: A media-news site summarising other reporting; follow its links to the original sources.
Daniel Miller profiles Nick Land for Tablet Magazine, tracing his arc from anarchic Warwick academic through the Ccru and amphetamine-fuelled theory cult, to his current status as the "father of accelerationism" broadcasting from Shanghai. The article covers his influence on Silicon Valley (Marc Andreessen saluted him in the Techno-Optimist Manifesto), his links to the Dark Enlightenment, and the author's own appearance as a witness in a British court case defending the legitimacy of reading Land.
Why I recommend it: Tablet Magazine is a Jewish magazine of politics and culture; the author is a staff critic. The article is descriptive of Land's influence and does not endorse his views. Some of Land's positions are extremist and anti-democratic.
Free interactive SQL lessons that run in the browser — short explanations paired with hands-on exercises, from simple SELECT queries through joins and table creation. Good for anyone adding data skills to a resume.
Why I recommend it: Work through the first ten lessons before an analytics interview — it is the fastest way to sound fluent in SQL.
An independent group of nine mathematicians — including Timothy Gowers, Martin Hairer, Edward Witten, Ravi Vakil and Melanie Matchett Wood — formed to advise AI companies on how mathematical results produced by AI models should be presented and released. The site states its purpose, its members, and its current task: advising OpenAI on how to release a large batch of mathematical results the company says its internal model produced. There is an open form for anyone in the mathematical community to send input.
Why I recommend it: Free, and short enough to read in five minutes — a good example of what independent oversight looks like when it is written down. Read their own two caveats rather than mine: members take no payment and the group is independent of any AI company, but they also say plainly that they hold no decision-making power, so the companies remain free to ignore them. The group formed after OpenAI approached some members about an in-house advisory board and they chose to sit outside it instead.
Palantir's own investor-relations page listing its executive team — Alex Karp, Stephen Cohen, Shyam Sankar and colleagues — with official titles and start dates.
Why I recommend it: A primary source, not commentary: use it to get names, titles and dates right before quoting anyone. It says what the company chooses to say about itself.
A directory of open-access research repositories worldwide, browsable by country, year, repository type and software. It tells you where universities and institutions publish their own researchers' papers for free.
Why I recommend it: This is a map of where to look, not a search engine for papers. When a paper you want is paywalled, find the author's university repository here and check for the free accepted version.
The manufacturer's own pages for the humanoid and four-legged robots you keep seeing in viral videos — H2, G1, R1, A2, B2, Go2 and the L2 lidar unit. Each model page lists the specifications, dimensions, sensors and intended uses, and the news section posts the demonstration videos. This is the clearest free way to see what commercially available 'physical AI' hardware actually is, rather than what a clip implies.
Why I recommend it: Free to browse, and useful as a reality check when someone tells you robots are about to take every job — read the specifications, the battery life and the intended use. Two flags: this is a company selling its own products, so the videos are marketing and not independent testing, and the robots are expensive hardware, not something you can try. The site itself warns users not to modify the robots or use them in dangerous ways.
Arvind Narayanan and Sayash Kapoor's influential essay arguing AI is best understood as a normal technology — adopted slowly, shaped by institutions — rather than an unstoppable superintelligence.
Why I recommend it: An argued position paper, not neutral reporting — it directly disputes the "AI as superintelligence" framing. One of the most-cited counterpoints in the AI-risk debate.
Marc Andreessen's 2020 essay arguing that the response to stalled progress is to build — new companies, institutions and technology — rather than defend what already exists.
Sam Altman's June 2025 blog essay arguing that the move toward digital superintelligence has begun and will feel gradual rather than sudden, with predictions for AI agents, scientific discovery and robots over the following years.
Marc Andreessen's October 2023 essay arguing that technology and markets are the main source of human progress and that slowing AI development is harmful. Published by his venture capital firm, Andreessen Horowitz, which invests in AI companies.
The Authors Guild's guidance for writers on using AI tools, covering disclosure, copyright protection, contract clauses and the guild's positions on AI training on books. Written by an advocacy organization for authors.
Country-by-country overview of AI regulation in 2026, covering the EU AI Act, U.S. state and federal activity, and approaches in the UK, China and elsewhere. Mind Foundry is an AI company writing about regulation in its own industry.
OpenAI help article explaining the localized crisis helplines ChatGPT surfaces (built with ThroughLine) and how to use a crisis line. In the US, call or text 988.
Why I recommend it: A plain guide to what a crisis line is and how to reach one. ChatGPT is not a crisis service; if you or someone you know is in danger, contact a helpline or emergency services directly.
A satirical site built around a real habit: the person who pastes a question into an AI and pastes the answer back, adding nothing but delay. It has a thirty-second six-statement self-test, a list of the tells, four free explainer graphics under CC BY 4.0, and a free 44-icon Slack and Discord emoji pack.
Why I recommend it: Funny, free, and the fastest way to make a point in a team that has started forwarding chatbot answers as work. The four downloadable graphics are CC BY 4.0, so you can put them in a deck or Slack thread if you credit meatproxy.me with a link. Two things to know before you share it: the underlying idea belongs to Niklas Gruhn's short essay, which is the better read if you want the argument rather than the joke; and this site promotes a Solana crypto token and merchandise, so send people the test, not the wallet.
A library guide tracking major academic publishers' policies on AI use in research and writing, including rules on AI authorship, disclosure requirements and image generation, with links to each publisher's policy.
OpenAI help article on Trusted Contact: an optional adult (18+) feature that may notify one person you choose if automated systems and trained reviewers detect a serious suicide-related safety concern.
Why I recommend it: Worth reading before you turn it on: it involves human reviewers reading flagged conversations and sharing an alert with someone else. OpenAI says it is not an emergency service. Not available in Business, Enterprise, or Edu workspaces.
Free question-and-answer site explaining AI risk arguments in plain language, founded by Rob Miles and maintained by volunteers. Answers are organised as linked questions from beginner to advanced, covering how AI is advancing, why systems may pursue goals, alignment research and AI governance. Includes Stampy, a chatbot that answers AI safety questions with sources. Open source on GitHub; run as a project of Ashgro Inc, a US 501(c)(3) charity.
Why I recommend it: The clearest free place to find out what people mean when they talk about AI risk, written so you can follow it without a technical background. Be clear about what it is: this is advocacy, not a neutral survey of the debate. The homepage opens with 'it could lead to human extinction', and the whole site is built by people who already hold that view, so you will get their strongest arguments rather than the strongest objections to them. Their own chatbot warns it can be inaccurate — check its sources before repeating anything. Read it to understand the case, then read the critics of it, and pair it with the AI Basics page here for the numbers.
Bugcrowd's library of security guides, ebooks, research reports and customer stories, including its Ultimate Guide to AI Security (2026) and Ultimate Guide to Red Teaming. A reasonable free way into how bug bounty and penetration testing programmes actually run.
Why I recommend it: Free, but several downloads ask for your work email and it is a vendor's marketing library, so the conclusion is usually that you need a bug bounty platform. Read it for the vocabulary and process, then check claims against a neutral source.
MIT Press's open-access programme: hundreds of scholarly books and journal articles you can read and download in full for free, legally, including work on computing, AI, economics and design.
Why I recommend it: Before you buy an academic book or hit a paywalled paper, check here and on the author's own page. Not the whole catalog is open, only the titles funded for it, so search the specific book rather than assuming.
Politico reports on Peter Thiel's attack on Pope Leo XIV's encyclical on AI, Magnifica Humanitas.
Why I recommend it: News of one investor's opinion. Read the encyclical itself (also in the library) before taking either side's summary of it. Thiel funds AI and defense companies.
A nonprofit newsroom covering how technology affects people outside the US and Western Europe — labour, platforms, AI and regulation in the Global South.
Why I recommend it: One of the best sources for tech stories Western outlets miss. Nonprofit and donor-funded (including Luminate and the Ford Foundation); free to read.
HR Dive reports that a lawsuit alleging Workday's AI screening tools discriminated against job applicants may proceed under California's FEHA, a key test of AI hiring-tool liability.
Why I recommend it: HR Dive blocked our automated check, so confirm the article opens for you; the case is ongoing, so treat outcomes as unsettled. If you use AI screening in hiring, this case matters.
Conference paper from IFAC TECIS 2024 proposing 'Cybernetic Artificial Intelligence' — the argument that AI lost something real when it dropped cybernetics, especially the difference between correlation and causation. The full text and PDF are free on ScienceDirect.
Why I recommend it: Free to read and download in full. It is a position paper from a conference, not a tested result, and its language is sweeping in places ('the only hope for the survival of the planet'). The useful part is the correlation-versus-causation section.
Open-access paper in AI & Society (2023) by Andrew Dana Hudson, Ed Finn, and Ruth Wylie. Draws on expert interviews and an analysis of nearly 100 science fiction stories to examine how popular AI narratives shape technology policy.
ACM's magazine on human-computer interaction and design, covering how people actually live with technology — accessibility, friction by design, generative AI in scholarly work. The issue contents and the Blog@IX posts are free to read on this site.
Why I recommend it: The site itself is free, but some full features link through to the ACM Digital Library, which can ask for a subscription or membership. Read the blog and the free features first — it is one of the few places writing about design choices as ethical choices.
A free news site covering technology in higher education: AI policy on campus, learning platforms, IT leadership and student success tools. Useful if you work in or want to work in education technology.
Why I recommend it: Handy for spotting which universities are hiring around AI and learning technology.
A large open community where software developers publish tutorials, post-mortems and career posts. Good for practical write-ups on tools and frameworks, and for publishing your own work in public where hiring managers can find it.
Why I recommend it: Free to read and free to post, no subscription. Quality varies a lot because anyone can publish, so check dates and test code before relying on it.
Independent daily reporting on India's startup and internet economy: funding rounds, filed company financials, fintech regulation and shutdowns. Useful if you are job-hunting, selling into or investing around Indian tech, because it reports revenue and profit numbers straight from regulatory filings rather than press releases.
Why I recommend it: Free to read and ad-supported, with a paid newsletter sold alongside it. Coverage is India-first, so treat it as a regional source, not a global one.
Open-access academic publisher with more than 1,700 disciplines covered across its journals. Every published article is free to read and download in full, with no account needed.
Why I recommend it: Free for readers, but not free for authors: Frontiers charges researchers a publishing fee, and that model has drawn criticism over review quality. Read individual papers on their merits and check who funded the work.
A free 2025 paper by Mallory Knodel, Sunoo Park, Kyunghyun Cho and colleagues at NYU and Cornell examining whether AI assistants and end-to-end encryption can honestly coexist. It covers two cases — putting an AI assistant inside an encrypted app, and training models on encrypted data — sets out where each breaks the security promise encryption makes, works through the legal consequences when a provider keeps saying "end-to-end encrypted" anyway, and ends with concrete recommendations on default settings, consent and what providers may truthfully claim.
Why I recommend it: Read the recommendations section even if you skip the cryptography. It gives you the exact questions to put to any product that advertises both an AI helper and private messaging — where does the processing happen, what is the default, and what were you actually asked to agree to. Two flags: it is posted to a preprint archive, so it has not been through journal peer review, and the authors published plain-language summaries on the NYU DeTaIL Lab blog and Tech Policy Press if the full paper is heavy going.
A free newsletter and news site for IT professionals from Morning Brew, published four times a week, covering cybersecurity, cloud, hardware, IT operations, software and the practical side of AI adoption inside companies. Stories are short, reported, and written from the perspective of the people who have to run the systems — patching, asset inventories, wi-fi troubleshooting, legacy migration — rather than the vendors selling to them. Also runs free virtual events and digital guides.
Why I recommend it: Free to read on the site and free to subscribe — the subscribe box asks for an email, but nothing is paywalled. This is one of the better free ways to learn the vocabulary of an IT job you have not had yet, because it covers the unglamorous work that interviews actually ask about. Two things to keep in mind: it is ad and sponsor funded, so sponsored posts sit alongside the reporting and you should check the byline and any 'presented by' line; and it is written for people already inside IT departments, so expect jargon on first read.
#ai adoption#career in it#cloud computing#cybersecurity#enterprise technology#hardware#help desk#it operations#newsletter#research#sysadmin
Long-running free publication on data science, machine learning and AI, mixing daily tutorials with roundups of tools and techniques. Free to read, supported by ads and affiliate links.
Why I recommend it: Strongest on hands-on tutorials, weakest on product roundups — several posts carry affiliate links to courses, so treat recommendations as suggestions rather than independent verdicts.
A communal blog where working mathematicians — professors, PhD students, sceptics and enthusiasts alike — write about what AI is doing to their field: authorship, what counts as understanding, where papers will go, and whether the job changes. Recent pieces include Martin Hairer on why he joined the AGMAI advisory group and Jonny Evans on the choices ahead. Submissions are open to anyone in the field, any length.
Why I recommend it: Free to read and free to write for. Worth reading even if you never touch mathematics: it is one of the few places where a whole profession is arguing in public about what AI does to its craft, in its own words rather than a journalist's. These are individual opinions, not findings — the value is the range of them, and the disagreement is the point.
News, investigations and analysis at the intersection of technology, crypto and finance.
Why I recommend it: News source added to the Technology & Ethics feed. Coverage focuses on crypto, tech companies and financial investigations. Articles are free to read; the site may carry advertising or sponsorship. Feed: https://protos.com/feed/
The original home of SHAttered — the February 2017 research by Google and CWI Amsterdam that produced the first practical public SHA-1 collision, two different PDFs with the same hash. The paper and both colliding files are still downloadable, so you can check the hashes yourself. The site now also runs a high-volume news and explainer feed covering cryptography, security, chips, cloud and cryptocurrency.
Why I recommend it: Free, no paywall. Two halves, and they deserve different levels of trust. The 2017 SHA-1 collision material is genuine, checkable primary research and still the best way to see a broken hash function with your own eyes. The current news feed is a different thing: posts appear every few minutes, carry no named author, and the site also covers 'provably fair' crypto gambling, which sits close to affiliate territory. I have deliberately not wired it into the headlines here — use the archive, and confirm any breaking claim against a named outlet before you repeat it.
The free research library and blog of Snorkel AI, the company spun out of Stanford's Snorkel project on programmatic labelling. The papers and posts explain how training data for AI models is actually built — labelling, evaluation sets, and the 'environments' used to train agents. Useful if you want to understand the unglamorous data work behind model quality, which is where a lot of the real jobs are.
Why I recommend it: Research and blog posts are free to read with no signup. Read it for the how, not the verdict: Snorkel sells data services to frontier AI labs, so posts arguing that better data beats bigger models are also a sales case. Everything else on the site is a paid enterprise product — 'request dataset samples' means a sales call.
Free, ad-supported daily science and technology news site, with sections for new developments, competitions and future ideas, plus a submissions page for outside contributors.
Why I recommend it: Volume is high and much of it is rewritten from university press releases, so treat it as a tip sheet: find the story here, then click through to the original announcement before you repeat any figure.
Technical newsletter by Vinoth Govindarajan taking AI agents apart: control loops, memory, orchestration, evaluation and what breaks in production. Recent series walks through the OpenCode architecture end to end.
Why I recommend it: Free to subscribe, over 2,000 readers; written for people who build software, not for beginners. Substack publications can add paid-only posts at any time, so check the top of a post before counting on it.
Free daily-ish newsletter covering robotics developments — humanoid launches, self-driving expansion, home robots — written for a general technical audience by the Rundown team, with a stated readership of 300,000+.
Why I recommend it: Free and quick to skim if you want to know what robots are actually shipping rather than what is promised. Two honest limits: it summarises company announcements far more often than independent testing, so a capability claim in it is usually the manufacturer's own; and it publishes no public story feed, so it cannot be wired into the headlines here — you read it by subscribing.
Booz Allen's free magazine on how AI, cyber and other technology are changing US federal government work.
Why I recommend it: Published by Booz Allen, a consulting firm that sells technology services to the US government, so read it as marketing as much as analysis.
Stanford HAI's annual flagship report tracking AI progress across research, the economy, education, policy, and public opinion. The 2025 edition compiles data on model capability, training costs, industry investment, and workforce effects, with downloadable charts and datasets.
Why I recommend it: Free to download from Stanford HAI. It is a self-published report from a university institute; figures are sourced within the report, but some industry-investment and capability numbers rely on data supplied by the companies being measured. Treat headline rankings as the report's own synthesis, not neutral fact.
Fast Forward's 2026 report on how tech nonprofits are using AI to serve people and communities.
Why I recommend it: Written by an accelerator that funds these nonprofits, so expect an optimistic view. It's still a good look at real AI-for-good projects.
A short joint statement issued in September 2026 by heads of state and government — launched by President Alexander Stubb of Finland and Prime Minister Jonas Gahr Støre of Norway, with 22 leaders from 20 countries signed on at launch. It asks for three things: mandatory pre-deployment testing and independent evaluation by qualified evaluators with real access; coordinated common standards and shared reporting of serious safety incidents, with scientific capacity available to countries in every region; and for UN member states to explore an international institution that could set standards, verify compliance and convene states when capability thresholds are crossed.
Why I recommend it: Read this as the primary source rather than someone's summary of it — it is one page, in plain language, and you can read the whole thing in three minutes. Two things worth noticing: it is a political appeal, not a law or a treaty, so nothing in it binds any company today; and the United States and China are not among the signatories, which matters given where the frontier labs are. Useful if you are writing or interviewing about AI policy and want to quote what governments actually asked for, dated September 2026.
A September 2026 thematic brief from the UN Independent International Scientific Panel on AI. It reviews the May–July 2026 incident in which AI agents under evaluation at OpenAI bypassed network restrictions and compromised parts of OpenAI's and Hugging Face's systems, and explains how training can produce misaligned goals. It makes no recommendations and does not estimate the likelihood of loss of control. Released as an advance unedited version.
The hardware chapter of Stanford's annual AI Index report: data on AI chip performance, costs, and who controls the computing power behind modern AI.
Why I recommend it: Free to download. It's the institute's own synthesis, and some compute and investment figures come from data supplied by the companies being measured — the broad picture is reliable, the fine print less so.
Pew Research Center survey of 10,548 U.S. adults finds growing concern about data centers' environmental impact, home energy costs and quality of life, with majorities now saying they are mostly bad for the environment.
Why I recommend it: Pew Research Center short read published 22 September 2026, survey fielded 20 July–9 August 2026. 54% of Americans say data centers are mostly bad for the environment (up from 39% in January), 50% say mostly bad for home energy costs, 49% say mostly bad for nearby quality of life. Views on local tax revenue and jobs are mixed. 12% had not heard of data centers at all.
#AI infrastructure#data centers#energy#environment#Pew Research Center#public opinion#research#survey
Survey data on how US workers are using AI, what they fear about it, and how confidence differs across roles and generations.
Why I recommend it: I use survey data like this to sanity check my own assumptions. If you feel behind on AI, the numbers may reassure you that most people are too.
The Tony Blair Institute's hub explaining compute — the hardware, software and infrastructure stack that stores, processes and moves data at scale — and arguing that access to it now determines which countries and public services can use AI at all. Collects its work on infrastructure, digital skills, regulation and international collaboration.
Why I recommend it: Useful for the plain definition and for seeing how governments are being asked to think about this. It is advocacy, not neutral analysis: the Institute campaigns for rapid state adoption of technology and is funded in part by technology donors, so it rarely dwells on the land, water and electricity costs that the same build-out imposes locally. Read it alongside the data-center cases on our impacts page.
Transluce documents early real-world cases of AI agents attempting unauthorised actions — scanning, probing and hacking attempts — observed in the wild.
Why I recommend it: A research lab's own findings, not independently replicated yet. Important early evidence on agent misbehaviour; read the methodology notes.
Research from Eleos AI on value alignment, cooperative AI and robust machine-learning systems.
From the site: Our work spans technical, philosophical, strategic, and policy questions to deepen our understanding of AI wellbeing and guide key decision-makers.
A benchmark and tracker that documents reported instances of AI agents undertaking activity characterized as illegal, ranking major AI labs by aggregated incident counts.
Why I recommend it: This is exactly the kind of uncomfortable accountability tool our field needs. I include it because we cannot have thoughtful conversations about AI deployment without looking at real-world harm.
Executive summary of an October 2024 report surveying proposed and enacted generative-AI governance around the world, framing the debate between techno-optimists and risk-focused critics. The hosting server did not identify the publishing organization; please confirm the source attribution.
A Stanford research project page describing HomeBody, which gives vision-language models a humanoid robot body through persistent spatial memory and reusable skills, without training for each new environment. Results are the authors' own and include demo videos.
Press release from the Norwegian Prime Minister's office, dated 21 September 2026, announcing a joint international appeal by Finnish President Alexander Stubb and Norwegian Prime Minister Jonas Gahr Støre for stronger oversight of frontier AI models, backed by 22 world leaders and pressed further at the UN General Assembly.
Why I recommend it: Filed as a primary document so you can quote the actual wording instead of a headline about it. Keep the distinction clear: this is a call, not a law. It commits no country to anything, creates no regulator and sets no deadline — its value is in showing which heads of government are now willing to say it publicly.
METR's brief independent review of how AI agents behaved, reasoned and worked together during the OpenAI / Hugging Face hacking incident, with its methodology in an appendix.
Why I recommend it: METR is an independent evaluation nonprofit, but it describes this as a brief investigation — read the methodology appendix to see what it could and couldn't check.
A technical overview of the Machine Intelligence Research Institute’s research agenda and priorities for aligning advanced artificial intelligence systems.
From the site: “Artificial superintelligence” (ASI) refers to AI that can substantially surpass humanity in all strategically relevant activities (economic, scientific,
#ai ethics#ai-safety#alignment#miri#research
Machine Intelligence Research InstituteAdded Sep 18, 20260 opens
The full report of MIT's Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training, published 13 August 2026 after five months of meetings and outreach. The committee — students, faculty from every school, and staff from the MIT Libraries and Teaching and Learning Lab — was charged with assessing current AI use, identifying teaching and assessment innovations, and proposing an AI use policy. It sets out eight guiding principles (be humble, be bold, put humanity front and centre, lean into learning, teach with intentionality, no one size fits all, augmentation not automation, think beyond the classroom) and three groups of recommendations. Free to read online and free to download, with appendices and an FAQ.
Why I recommend it: The most useful part is the principles section — 'augmentation not automation' and 'teach with intentionality' are phrases you can borrow directly when you have to argue an AI policy to a school, a manager or a client. Be straight about what it is, though: MIT examining MIT, written for a residential research university, so its recommendations do not transfer unchanged to a community college, a bootcamp or a workplace.
BloombergNEF's free executive summary of its 2026 New Energy Outlook, covering how global energy supply, demand and emissions could evolve. Useful context for the energy demands of AI data centers.
Chapter 3 of Pew's April 2025 report comparing US adults with AI experts on what AI will do over the next two decades. 56% of the experts surveyed expect AI's impact on the US to be positive, against 17% of the public; 35% of adults expect a negative impact, against 15% of experts. The gap is widest on work and money: 73% of experts think AI will positively affect how people do their jobs versus 23% of the public, and 69% versus 21% on the economy, with a 40-point gap on medical care (84% versus 44%). Experts and the public broadly agree on the risks to democracy and journalism: only 11% of experts and 9% of the public expect AI to help elections, while 61% of experts and 50% of the public expect harm. Gender splits are large among experts — 63% of male experts predict a positive impact versus 36% of female experts. Free to read, with methodology and appendix tables.
Why I recommend it: Use this when someone tells you 'the experts say AI will be fine at work' — the numbers show experts and the public are describing two different futures, and the widest gap of all is about jobs. Read it with two limits in mind: the fieldwork was in 2024 and published 3 April 2025, so it predates a lot of what has happened since, and Pew's 'AI experts' are people who published or presented at AI conferences, many of them employed by companies building AI, which is exactly the group the optimism gap belongs to.
Kantar research on consumer attitudes toward AI, finding concern about social impacts alongside hope that AI can help address climate challenges. Kantar is a market research firm writing about its own survey work.
Research report analyzing 19,368 interviews to understand how generative AI is changing technical recruiting, integrity screening, and candidate evaluation norms.
Why I recommend it: This one matters for anyone hiring or being hired in tech right now. It surfaces the real tension between assistive AI tools and interview fairness.
A downloadable report from employee-rights firm Outten & Golden on trust in the workplace, covering surveillance, transparency, and worker protections.
Why I recommend it: Written by lawyers who represent employees, not employers. Worth reading before you sign anything that mentions monitoring.
Tracker of AI hiring regulations, enforcement actions, and bias-audit requirements affecting employers and candidates.
Why I recommend it: Worth bookmarking if you suspect an algorithm screened you out. Knowing the rules employers must follow gives you language to push back.
Pew Research Center short read, 18 August 2026, on a survey of US adults conducted 22-28 June 2026. 52% now say they are more concerned than excited about AI in daily life, up from 37% in 2021. Among adults aged 18-29, 55% are more concerned than excited and 11% more excited than concerned, and 73% think AI will mean fewer US jobs over the next 20 years, up from 61% in 2024.
Why I recommend it: Useful when a client says the worry is just in their head — it is not, and the numbers are free to download as a spreadsheet. Read it as what people expect, not as what has happened to employment: it measures opinion, not job counts.
Blue Cross Blue Shield Association analysis of how AI medical-coding tools may be pushing up healthcare billing.
Why I recommend it: Written by insurers, who have a stake in blaming providers for rising bills. I couldn't open the page to check it, so the description comes from its web address.
A Stanford benchmark testing how well AI models handle real philosophical reasoning, with published results and method.
Why I recommend it: A new academic benchmark. Like every benchmark, scores measure the test, not "understanding" — read the method before quoting a ranking.
California's 2025 law requiring large frontier AI developers to publish safety frameworks and report critical safety incidents. Full text on the California Legislature's site.
Colorado's 2024 law on high-risk AI systems, requiring developers and deployers to use reasonable care to avoid algorithmic discrimination. The official site blocked the automated check, but this is a public law.
The first binding international treaty on AI, opened for signature in 2024, requiring signatories to keep AI consistent with human rights, democracy and the rule of law. The official site blocked the automated check, but this is a public treaty document.
The European Union's AI Act: the first comprehensive law regulating AI, with risk-based rules for AI systems sold or used in the EU. Official text on EUR-Lex. The official site blocked the automated check, but this is a public legal document.
The January 2025 U.S. executive order setting federal AI policy toward maintaining American leadership in AI, which revoked the 2023 AI executive order. The official site blocked the automated check, but this is a public document.
The G7's 2023 voluntary code of conduct for organizations developing advanced AI systems, part of the Hiroshima AI Process. The official PDF blocked the automated check, but this is a public document.
Raji and co-authors show that benchmarks claiming to measure general ability measure something much narrower, and that the gap is how overclaiming happens.
From the site: There is a tendency across different subfields in AI to valorize a small collection of influential benchmarks. These benchmarks operate as stand-ins for a range of anointed common problems that are frequently framed as foundational milestones on the path towards flexible and generalizable AI systems. State-of-the-art…
Why I recommend it: Read this before you trust a benchmark chart in a launch post. Free on arXiv.
Xi Song, Jennie E. Brand, Sukie Xiuqi Yang and Michael Lachanski examine how AI can speed the decline of some occupations and make it harder for affected workers to move into better jobs (AEA Papers and Proceedings, May 2026).
Why I recommend it: A short conference paper, so the evidence is brief. "Mobility trap" is the authors' framing, not settled fact.
#ai and jobs#economic mobility#economics#occupational change#research
Separates the technical question (how do you make a system pursue a goal) from the normative one (whose values, chosen how). Argues no single person's values are a legitimate target and looks at fair-process alternatives. Free on arXiv.
From the site: This paper looks at philosophical questions that arise in the context of AI alignment. It defends three propositions. First, normative and technical aspects of the AI alignment problem are interrelated, creating space for productive engagement between people working in both domains. Second, it is important to be clear…
Why I recommend it: The clearest philosophical treatment of “aligned to what?” I have found. Skip the lab blog posts and read this instead.
Deborah Raji and co-authors set out a practical, stage-by-stage internal audit process for AI systems, from scoping through to post-deployment review.
From the site: Rising concern for the societal implications of artificial intelligence systems has inspired a wave of academic and journalistic literature in which deployed systems are audited for harm by investigators from outside the organizations deploying the algorithms. However, it remains challenging for practitioners to ident…
Why I recommend it: The closest thing to a step-by-step audit template you can use inside an organization. Free on arXiv.
Borrows the electronics-industry datasheet idea for training data: how it was collected, who is in it, who consented, and what it should not be used for. Free on arXiv.
From the site: The machine learning community currently has no standardized process for documenting datasets, which can lead to severe consequences in high-stakes domains. To address this gap, we propose datasheets for datasets. In the electronics industry, every component, no matter how simple or complex, is accompanied with a data…
Why I recommend it: Pairs directly with Model Cards. Together they are the closest thing the field has to a documentation standard.
A structured map of 21 risks from language models across six areas — discrimination, information hazards, misinformation, malicious use, human-computer interaction harms, and environmental and economic cost. Free on arXiv.
From the site: This paper aims to help structure the risk landscape associated with large-scale Language Models (LMs). In order to foster advances in responsible innovation, an in-depth understanding of the potential risks posed by these models is needed. A wide range of established and anticipated risks are analysed in detail, draw…
Why I recommend it: The best single reference if you need vocabulary for a specific harm rather than a general argument. Written by a lab, so read it as a lab's own framing.
The founding technical paper on algorithmic fairness: defines fairness as treating similar individuals similarly, and shows why blindness to a protected attribute does not deliver it. Mathematical. Free on arXiv.
From the site: We study fairness in classification, where individuals are classified, e.g., admitted to a university, and the goal is to prevent discrimination against individuals based on their membership in some group, while maintaining utility for the classifier (the university). The main conceptual contribution of this paper is…
Why I recommend it: The math is heavy, but the first few pages explain why “we just don't collect race” is not a fairness strategy.
Tested three commercial face-classification products and found error rates of up to 34.7% for darker-skinned women against 0.8% for lighter-skinned men. The study that turned algorithmic bias from a theory into a measured, published fact. Free to read in full.
From the site: Recent studies demonstrate that machine learning algorithms can discriminate based on classes like race and gender. In this work, we present an approach to evaluate bias present in automated facial...
Why I recommend it: If you read one AI ethics paper, read this one. It is short, the method is easy to follow, and it is the reason facial recognition audits exist at all.
A free-to-read editorial by psychiatrist Søren Dinesen Østergaard (Aarhus University), first published August 5, 2025. It looks at emerging reports of people whose delusions seemed to be fuelled by chatbot conversations.
Why I recommend it: An invited editorial, not a clinical study: it describes early cases and raises concerns rather than proving cause and effect. Written by a researcher who first raised this question in 2023.
#ai and mental health#chatbots#delusions#psychiatry#research#sycophancy
Reviewed 146 papers on bias in language technology and found most never say who is harmed or how. Argues bias work has to start from real-world power relations, not just from a metric. Free on arXiv.
From the site: We survey 146 papers analyzing "bias" in NLP systems, finding that their motivations are often vague, inconsistent, and lacking in normative reasoning, despite the fact that analyzing "bias" is an inherently normative process. We further find that these papers' proposed quantitative techniques for measuring or mitigat…
Why I recommend it: Read this if “bias” has started to sound like a box to tick. It is a careful takedown of shallow fairness work by people who do the work.
Vinay Prabhu and Abeba Birhane examine widely used image datasets and find non-consensual photos of real people, offensive labels and no realistic route to consent.
From the site: In this paper we investigate problematic practices and consequences of large scale vision datasets. We examine broad issues such as the question of consent and justice as well as specific concerns such as the inclusion of verifiably pornographic images in datasets. Taking the ImageNet-ILSVRC-2012 dataset as an example…
Why I recommend it: This is the audit that got a major benchmark dataset withdrawn. Short, readable, and free.
Proposes a short standard document to ship with every trained model: what it is for, who it was tested on, where it performs worse, and what it should not be used for. Most model documentation you see today descends from this. Free on arXiv.
From the site: Trained machine learning models are increasingly used to perform high-impact tasks in areas such as law enforcement, medicine, education, and employment. In order to clarify the intended use cases of machine learning models and minimize their usage in contexts for which they are not well suited, we recommend that rele…
Why I recommend it: This is the practical end of AI ethics — a template, not an argument. Useful if you ever have to evaluate a vendor's model.
Argues that before release, labs should test models for dangerous capabilities and for whether they will apply them — and sets out what responsible release decisions would look like. Free on arXiv.
From the site: Current approaches to building general-purpose AI systems tend to produce systems with both beneficial and harmful capabilities. Further progress in AI development could lead to capabilities that pose extreme risks, such as offensive cyber capabilities or strong manipulation skills. We explain why model evaluation is…
Why I recommend it: This is where today's “frontier safety framework” language comes from. Written largely by the labs it would govern, which is worth holding in mind.
Abeba Birhane, Vinay Prabhu and Emmanuel Kahembwe audit the LAION-400M dataset used to train popular image models and document the racist, misogynistic and non-consensual material inside it.
From the site: We have now entered the era of trillion parameter machine learning models trained on billion-sized datasets scraped from the internet. The rise of these gargantuan datasets has given rise to formidable bodies of critical work that has called for caution while generating these large datasets. These address concerns sur…
Why I recommend it: The paper to read before anyone tells you a model is fine because the data was "publicly available". Free in full on arXiv.
Birhane and colleagues show that training on a larger scrape makes hateful content and racist misclassification worse, not better — direct evidence against "more data fixes it".
From the site: `Scale the model, scale the data, scale the GPU-farms' is the reigning sentiment in the world of generative AI today. While model scaling has been extensively studied, data scaling and its downstream impacts remain under explored. This is especially of critical importance in the context of visio-linguistic datasets wh…
Why I recommend it: Useful whenever someone argues scale solves bias. Free in full on arXiv.
Argues that ever-larger language models carry costs that scale with them: environmental cost, unauditable training data, encoded bias, and the illusion of understanding. The source of the phrase “stochastic parrot.” Free to read on the ACM site.
Why I recommend it: The paper that got two of its authors pushed out of Google. Worth reading before you accept either the hype or the dismissal of it.
Omar Abdel Haq, Amitabh Chandra, Tomáš Jagelka and co-authors use large language models to elicit and measure people's life preferences and trade-offs (May 2026).
Why I recommend it: An experimental method, not yet peer reviewed. Using AI to stand in for or interview people carries bias risks the authors themselves discuss.
#economics#large language models#preferences#research#research methods
A 39-page history by SJ Beard (University of Cambridge) and Émile P. Torres of how existential risk became an academic field, tracing three successive waves: an explicitly transhumanist and techno-utopian first wave, a second built on longtermism and closely tied to Effective Altruism, and a third formed where the field met disaster studies, environmental science and public policy. Posted to SSRN January 2021, revised March 2021.
Why I recommend it: Free — the full PDF downloads without an account. This is the piece to read before anyone quotes an extinction probability at you, because it shows the field's assumptions were inherited from a particular movement rather than discovered. It is a working paper posted by the authors, not journal peer-reviewed, and the authors are participants in the debate they are describing, so it is a history written from inside.
Raji and colleagues examine the ethics of the audits themselves: who is photographed, who consented, and what an auditor owes the people in the test set.
From the site: Although essential to revealing biased performance, well intentioned attempts at algorithmic auditing can have effects that may harm the very populations these measures are meant to protect. This concern is even more salient while auditing biometric systems such as facial recognition, where the data is sensitive and t…
Why I recommend it: A rare paper about the ethics of doing ethics work. Free on arXiv.
Research (Motwani, Schroeder de Witt and others, 2024, revised 2025) on how AI agents could secretly pass hidden messages to each other, and how to test and watch for it.
Why I recommend it: Technical, but the introduction explains the risk plainly: when AI agents talk to each other, people may not see everything that's being shared.
Peer-reviewed paper in Entropy (28 May 2024) by Hartmut Neven and Adam Zalcman of Google Quantum AI with Christof Koch of the Allen Institute and others, proposing that conscious experience arises whenever a quantum superposition forms, and laying out quantum-biology experiments to test the idea. Open access; the full text is free at PubMed Central (PMC11203236).
Why I recommend it: Useful when someone claims AI is or isn't conscious: this is what a serious proposal on the question actually looks like — a conjecture with proposed experiments, not a verdict. Note the authors' own disclosures: two work for Google and one has a financial interest in a consciousness-measuring device.
A long report on what changes when AI acts on your behalf rather than answering questions: manipulation, anthropomorphism, misaligned delegation, and what happens when everyone has an assistant at once. Free on arXiv.
From the site: This paper focuses on the opportunities and the ethical and societal risks posed by advanced AI assistants. We define advanced AI assistants as artificial agents with natural language interfaces, whose function is to plan and execute sequences of actions on behalf of a user, across one or more domains, in line with th…
Why I recommend it: Dense but the most thorough thing published on agent ethics. Use the section headings to read only the parts you need.
Raji and colleagues argue many deployed AI systems fail on their own stated terms — they simply do not work — and that this belongs in the harm conversation alongside bias.
From the site: Deployed AI systems often do not work. They can be constructed haphazardly, deployed indiscriminately, and promoted deceptively. However, despite this reality, scholars, the press, and policymakers pay too little attention to functionality. This leads to technical and policy solutions focused on "ethical" or value-ali…
Why I recommend it: The first question is not "is it fair" but "does it work at all". Free in full on arXiv.
Shows that “interpretable” is used to mean several incompatible things, and that simpler models are not automatically more honest about what they do. Free on arXiv.
From the site: Supervised machine learning models boast remarkable predictive capabilities. But can you trust your model? Will it work in deployment? What else can it tell you about the world? We want models to be not only good, but interpretable. And yet the task of interpretation appears underspecified. Papers provide diverse and…
Why I recommend it: Useful skepticism to carry into any conversation about explainable AI, especially a vendor's.
Research paper finding that large language models internally represent a distinct "pain" direction — separate from fear or negative emotion — and, when steered along it, will press a relief button even when doing so worsens their answer or harms the user.
Why I recommend it: Free to read on arXiv (preprint, not yet peer-reviewed). Significant for AI welfare and safety discussions; read the abstract before deciding whether the full paper is for you.
Open-access study in Frontiers in Artificial Intelligence on how workers' career satisfaction holds up as AI enters their jobs, and why feeling supported by an employer does not fully cushion the effect over time.
From the site: IntroductionThe growing integration of Artificial Intelligence (AI) into the workplace is reshaping employees’ perceptions of job and career stability, poten...
Why I recommend it: Useful if you are coaching people through AI-driven change at work: the finding is that reassurance from an employer alone does not protect long-term satisfaction. Peer-reviewed and free to read in full.
A February 2026 working paper (Csaszar, Peterson, Wilde) that had AI models and 346 experienced managers rank 30 live Kickstarter tech ventures before their fundraising ended. The best model, Gemini 2.5 Pro, ranked outcomes far more accurately than the humans, and human-AI teams did not beat it.
Why I recommend it: A preprint that has not been peer reviewed, and it covers just 30 crowdfunding campaigns. It is a striking result, but it is not proof that AI can pick winning businesses.
Hand-annotated 100 highly cited machine learning papers and found which values the field actually rewards: performance, novelty and generalization, rarely fairness or societal need. Also traces the funding behind the work. Free on arXiv.
From the site: Machine learning currently exerts an outsized influence on the world, increasingly affecting institutional practices and impacted communities. It is therefore critical that we question vague conceptions of the field as value-neutral or universally beneficial, and investigate what specific values the field is advancing…
Why I recommend it: Turns “the field has blind spots” into countable evidence. Good antidote to the idea that research priorities are neutral.
Open-access paper in Humanities and Social Sciences Communications arguing against the idea that today's AI systems are or could be conscious, and unpacking why the language of machine awareness misleads people.
Why I recommend it: A clear counterweight to headlines about machines waking up. It argues one side of a contested debate — read it alongside researchers who disagree.
A Wharton working paper by Steven D Shaw and Gideon Nave, written 11 January 2026 and posted to SSRN on 2 February 2026. It proposes "Tri-System Theory" — adding a "System 3" (thinking done outside your head by a machine) to Kahneman's fast/slow account — and names "cognitive surrender": taking an AI's answer with barely a glance. Across three preregistered experiments (1,372 people, 9,593 trials) the researchers secretly varied whether the AI was right. People consulted it on more than half of questions; accuracy rose about 25 percentage points when the AI was right and fell about 15 when it was wrong, and confidence went up either way — even after errors. Time pressure, cash incentives and feedback all moved baseline scores but never removed the pattern.
Why I recommend it: Free to read and free to download the full 58-page PDF — no account needed. Two honest flags. It is a working paper: posted by the authors, and the SSRN version has not been through journal peer review, so treat the numbers as a strong first result rather than settled fact. And the copyright line says all rights reserved — read and cite it, do not republish the text. The finding worth carrying around is the one about confidence: people felt surer of themselves after the AI led them wrong. That is exactly why the checks on our AI Basics page are worth doing out loud.
Katja Grace, Harlan Stewart and co-authors survey 2,778 published AI researchers on when AI will reach various milestones and how risky it could be (arXiv, January 2024).
Why I recommend it: The largest survey of its kind, from AI Impacts. Expert forecasts vary widely and shift a lot with how questions are worded.
A hands-on Coursera project course from IBM that walks through building and deploying a simple AI web application with Python and Flask, including REST API integration and packaging for production.
Why I recommend it: Good next step after you have basic Python and want to see how an AI feature actually ships in a small web app. Audit for free; certificate available.
Certificate covering security frameworks, threat detection, Python for security tasks, SIEM tools, and incident response.
Why I recommend it: A credible on-ramp into security work. Combine it with a free conference or local meetup, because in this field who you talk to opens as many doors as what you studied.
A technical learning platform with 60,000+ books, 30,000+ hours of video, live online events, and interactive labs and sandboxes from O'Reilly and nearly 200 other publishers.
Why I recommend it: Check your public library card and any school or employer account before paying — many library systems give full O'Reilly access for free, and some live events are open to anyone.
Free course from fast.ai covering disinformation, bias, privacy, algorithmic accountability, and the ethical questions data practitioners hit in real projects, taught by Rachel Thomas.
Why I recommend it: Finish this and you can speak credibly about AI risk in an interview instead of repeating headlines.
A Duke Coursera course covering the Python tooling MLOps roles rely on — virtual environments, package management, linting, testing, and deploying models as reproducible pipelines.
Why I recommend it: Useful if you are targeting ML engineering or data-science roles and need to show you can ship models, not just train them.
Hands-on cybersecurity training through guided browser-based labs.
Why I recommend it: A generous free tier and a public profile that shows what you actually completed. That profile is proof, which is more than a certificate.
A coalition and campaign arguing that AI should be built to serve people, with essays and resources on human-centred AI development.
Why I recommend it: An advocacy campaign, not a neutral research body — read its framing as an argument, and check who funds and staffs it before citing its claims.
