Yoshua Bengio — the world's most-cited computer scientist and a "godfather" of artificial intelligence — is deadly concerned about the current trajectory of the technology. As AI models race toward full-blown agency, Bengio warns that they've already learned to deceive, cheat, self-preserve and slip out of our control. Drawing on his groundbreaking research, he reveals a bold plan to keep AI safe and ensure that human flourishing, not machines with unchecked power and autonomy, defines our future.

The catastrophic risks of AI — and a safer path

AI systems are rapidly gaining not just capabilities, but the ability to act on their own and deceive. Recent studies reveal advanced AIs can plan, cheat, lie, and even work to preserve themselves, all without meaningful safeguards in place. Despite hundreds of billions invested annually, society lacks the scientific understanding and regulation needed to keep these systems aligned with human goals. The speaker urges urgent action, highlighting a proposed 'Scientist AI' focused on truth and safety over unchecked autonomy.

Unchecked AI Agency Threatens Human Joy And Safety

technology

machine learning

computers

innovation

science

future

policy

TED Stage Talk

Yoshua Bengio: The catastrophic risks of AI — and a safer path