Yoshua Bengio
Turing Award laureate; chair of the International AI Safety Report
Professor at the Universite de Montreal and founder of Mila, who shared the 2018 Turing Award for deep learning. He now chairs the International AI Safety Report and founded the nonprofit LawZero.
Read them in their own words
Published elsewhere, not by me — so you get real context beyond the name and the title.
'It keeps me awake at night': machine-learning pioneer on AI's threat to humanity
Why one of the field's founders now spends his time on catastrophic risk. Nature may ask you to subscribe.
Key arguments & positions
- Publicly changed position around 2023, saying he had underestimated catastrophic risk from systems he helped invent.
- Argues for building "non-agentic" scientist AI that explains and predicts rather than pursuing goals in the world.
- Chairs the International AI Safety Report and pushes for independent evaluation of frontier models before release.
Accomplishments
- Shared the 2018 Turing Award with Geoffrey Hinton and Yann LeCun for deep learning.
- Founder and scientific director of Mila, the Quebec AI Institute; professor at the Universite de Montreal.
- Founded the nonprofit LawZero (2025) to build safety-focused AI; chairs the International AI Safety Report.
Papers & key writings
Links
In the library
Nothing of theirs is filed in the hub yet. Browse the full library.
Recommended next
Hand-picked from the hub based on what Yoshua Bengio covers.
Practical Deep Learning for Coders (fast.ai)
Top-down deep learning course using PyTorch, covering computer vision, NLP, and model deployment. Assumes existing Python basics.
Why this: Also about deep learning
Deep Learning Specialization (DeepLearning.AI)
Five-course series on neural networks, convolutional and sequence models, and how to structure ML projects.
Why this: Also about deep learning
AI as Normal Technology
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 this: Also about AI risk
Fool's Expertise
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.
Why this: Also about AI risk