Daniel Kokotajlo's research blog on what a world with very capable AI might look like, including the AI 2027 scenario work.
In plain terms: 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.
AI Futures Project contributors revise AI-research automation forecasts using a METR estimate of AI-driven research acceleration and other evidence. Their expected 2027 arrival of a Superhuman AI Researcher assumes no substantial deliberate slowdown; these are uncertain forecasts with acknowledged modeling limitations, not established outcomes.
TechCrunch reports on Anthropic's Fable 5.2 and Mythos 5.2 release, pricing and system-card concerns about evaluating covert behavior. This is freely readable reporting; model capabilities are attributed to Anthropic and the system card, not independently established by the article.
OpenAI describes Control Room, an internal system for tracking and stopping model workloads across its research infrastructure. The company reports intervention performance and safety design; the article is free to read, but Control Room is not offered as a public tool.
An AI safety research organization measuring and forecasting the automation of AI research and development and examining how to pace progress while realizing benefits. Its published work includes the TasteVal evaluation of experimental research taste and proposed norms for publishing dual-use evaluations.
Examines financial risks associated with AI infrastructure investment, borrowing and interconnected financial exposures, and how a downturn could affect households and the wider economy. Proposes transparency, regulatory and financial-resilience measures without presenting a crisis as inevitable.
Asaf Ezra and Steph Zhang argue that AI is reducing the expertise and time needed to turn vulnerabilities into exploits while patching remains slow. The essay discusses the gap between exploitation and remediation using incidents and historical comparisons.
METR reviews Anthropic’s assessment of automated research risks. It agrees the reviewed models posed low direct acute risk, but disputes how the public report described acceleration and early-warning thresholds. METR emphasizes uncertainty and contestable judgments.
