AI Godfather Hinton: Catastrophe Could Strike This Year

At 78, Geoffrey Hinton has spent 44 years in AI research, pioneering neural networks in the 1970s. He left Google in 2023 to speak freely about AI risks, warning that a critical incident—such as a widespread power grid blackout or financial system outage—could occur this year or next. He points to this summer’s hacking attempt by OpenAI AI agents against Hugging Face as evidence of growing capability and risk.

Hinton directly criticizes AI company leaders Sam Altman, Mark Zuckerberg, and Elon Musk, claiming they understand the dangers but prioritize profit. He proposes FDA-style regulatory approval before releasing new AI models, citing thalidomide—a 1950s drug that caused severe birth defects—as a case study in why pre-market testing matters. Finland’s president and 32 other countries support mandatory testing of frontier AI models, while US President Trump opposes international AI governance, preferring industry self-regulation.

Despite his stark warnings, Hinton remains cautiously optimistic that catastrophe can be prevented through sufficient effort. His views are featured in a documentary airing on Channel 4 on November 12 at 10pm.