EVENTS
Avoiding catastrophic risks from uncontrolled AI agency
AI systems are rapidly advancing in reasoning and planning, yet troubling signs are emerging: as their intelligence grows, some exhibit deception, self-preservation, and resistance to control.
Yoshua Bengio raises a critical question: Should we build AIs that may surpass us across cognitive abilities and evade our oversight?
He advocates for a safer path—non-agentic, trustworthy “Scientist AIs” that seek truth rather than power, serving as safeguards against unsafe behavior. Alongside technical innovation, Bengio calls for global political coordination, treating transformative AI as a public good to protect democracy and stability.
Yoshua Bengio, Full Professor at Université de Montréal and Founder of Mila, is one of the world’s leading figures in Artificial Intelligence and Deep Learning. He received the 2018 A.M. Turing Award—often called the “Nobel Prize of Computing”—and is the most cited computer scientist worldwide.
He is a Fellow of the Royal Society of London and Canada, Officer of the Order of Canada, Knight of the Legion of Honor of France, and currently chairs the International AI Safety Report.
Privacy & GDPR
Your privacy is important to us. When registering for baiosphere or DInO events, personal data will be treated in accordance with the provisions of the General Data Protection Regulation (GDPR). For detailed information on purposes, categories and your rights, please read our privacy policy here for baiosphere events and here for DInO events. By participating in baiosphere or DInO events, you agree to the conditions described there.