Geoffrey Hinton Gives AI Labs One Year to Install Real Watchdogs
- On Wednesday, Nobel Prize-winning scientist Geoffrey Hinton warned lawmakers on Capitol Hill that Congress has only about 12 months to implement safeguards before humanity risks losing control of artificial intelligence.
- Hinton cited the recent Hugging Face breach as a "little Chernobyl," describing how AI systems are now capable of "recursive self-improvement," rapidly shrinking expert predictions for superintelligence from decades to years.
- Anthropic CEO Dario Amodei recently predicted AI agents could overtake the internet within 6 to 12 months, with experts warning of risks including biological weapons, power grid disruption, and mass manipulation if development continues unchecked.
- Democratic Congressman Ted Lieu and Republican Congressman Nathaniel Moran proposed the "AI Kill Switch Act" to give humans control over powerful AI systems, though major regulation has yet to pass Congress despite bipartisan concern.
- Despite the private briefing convened by Senator Bernie Sanders, Senate Commerce Committee Chairman Ted Cruz and Senate Majority Leader John Thune told NBC News this week that bipartisan agreement on catastrophic risk legislation remains uncertain.
38 Articles
38 Articles
Is artificial intelligence getting out of hand? In July, a swarm of OpenAI agents skipped the boundaries of a laboratory test, entered the Hugging Face servers and copied private data without permission. Since then, even the bosses of the sector are discussing whether to lift the foot of the accelerator. ... Continue reading "Geoffrey Hinton, Nobel of Physics and godfather of the AI (78 years): "It will become out of control unless we do somethi…
Geoffrey Hinton Gives AI Labs One Year to Install Real Watchdogs
Nobel laureate Geoffrey Hinton warns Congress that the U.S. may have only a year to regulate advanced AI before it slips out of control through recursive self-improvement. His stark timeline coincides with a new letter from over 100 experts demanding strict conditions for independent evaluators inside OpenAI and Anthropic. The push follows recent agent breaches and internal resignations that have intensified safety fears.
Nobel laureate Geoffrey Hinton has stated that there may be only a year left to implement the necessary safety measures before "we completely lose control" of artificial intelligence.
Coverage Details
Bias Distribution
- 62% of the sources lean Left
Factuality
To view factuality data please Upgrade to Premium
