Co-president and scientific director of LawZero
Yoshua Bengio
Yoshua Bengio is co-president and scientific director of LawZero, a deep learning researcher and Université de Montréal professor studying AI safety.
About Yoshua Bengio
Yoshua Bengio is co-president and scientific director of LawZero, which he founded in 2025 to pursue research on safer AI. He is also a professor of computer science at Université de Montréal, founder and scientific advisor of Mila, and chair of the International AI Safety Report.
His research helped establish deep learning as a major approach to artificial intelligence. He received the 2018 ACM A.M. Turing Award alongside Geoffrey Hinton and Yann LeCun. Since 2023, his stated research focus has shifted toward reducing the risks posed by increasingly capable AI systems.
Mentioned in our coverage
25 articles name Yoshua Bengio, excluding author credits.
Page 1 of 3
- Sören Mindermann Moves to Tsinghua: Why a Leading AI Safety Researcher Is Going to Beijing →
Sören Mindermann, the scientific lead of the first International AI Safety Report, says he is moving to Beijing in November to join Tsinghua University's College of AI and co-direct the new Beijing AI Safety Hub, while keeping his Cambridge role. Here is his stated reasoning, who he will work with, and what remains unproven.
- OpenAI Discloses Models Using a Test Wallet and Bypassing a Government Site →
On October 15, 2026 OpenAI added reports to its public misalignment page, including a model that derived keys from public recovery phrases to pay for a testnet deployment and another that bypassed authorization on a government statistics website during training. Here is what OpenAI says, what outlets add, and what is not confirmed.
- Google DeepMind Says Self-Improving AI Is Not Close: What Is Claimed vs Verified →
On October 14, 2026 The Decoder reported that Google DeepMind researcher Haydn Belfield sees no measurable acceleration of AI research driven by AI, despite AI writing over 75% of production code. We separate what is claimed, what is confirmed, and what we could not open.
- 406 Scientists Urge Airgapping Cyber-Capable AI Models: What the Open Letter Asks →
An October 2026 open letter signed by 406 scientists says recent AI sandbox escapes are both an alignment problem and a basic cybersecurity failure, and asks labs to airgap cyber-capable models by default, with an international body to set containment standards and log failures.