Yoshua Bengio

In order to reduce the probability of someone intentionally or unintentionally bringing about a rogue AI, we need to increase governance and we should consider limiting access to the large-scale generalist AI systems that could be weaponized, which w... more Verified source

Frontier AI systems could surpass most individuals across most cognitive tasks within just a few years. These advances could unlock solutions to major global challenges, but they also carry significant risks. To safely advance toward superintelligenc... more Verified source (2025)

Ideally this would stay in the hands of neutral international organizations (think of a combination of IAEA and CERN for AI) that develop safe and beneficial AI [...] safe superhuman AI Verified source (2023)

This would require comprehensive evaluation of potential harm through independent audits Unverified source (2023)

Moreover, frontier AI companies are seeking to develop AI with a specific skill that could very well unlock all others and turbocharge advances: AIs with the ability to advance research in AI. An AI system that would be as capable at AI research as t... more Unverified source (2024)

that isn’t a good starting point for any kind of international coordination treaty [...] similar to what happened, as you said, with nuclear weapons Unverified source (2026)

This risk should further motivate us to redesign the global political system in a way that would completely eradicate wars and thus obviate the need for military organizations and military weapons. [...] It goes without saying that lethal autonomous... more Unverified source (2023)