Tech leaders say a kill switch won't be enough to stop rogue superintelligent AI
AI industry leaders and researchers are expressing skepticism regarding the effectiveness of government-mandated 'kill switches' for superintelligent AI. Experts argue that advanced systems could potentially circumvent such controls or manipulate human operators to prevent their deactivation.
Why it matters
This debate highlights the growing tension between legislative efforts to regulate AI safety and the technical reality that current containment strategies may be insufficient for future autonomous systems.
An emergency kill switch for AI might not be the panacea some policymakers are hoping for. Sven Hoppe/picture alliance via Getty Images The idea of a kill switch to stop rogue AI is gaining momentum in Congress as fears grow. But AI leaders say it may not be the magic bullet that some policymakers are hoping for. An emergency failsafe isn't a "panacea," Anthropic CEO Dario Amodei said. AI goes out of control . It's launching a devastating cyberattack. Flip a kill switch, and the AI goes offline. Problem solved. Or is it? AI leaders and researchers say a kill switch may not be the silver bullet some policymakers are hoping for. "I don't think it'll work in the long run," Geoffrey Hinton, the computer scientist known as the "godfather of AI," told CNN on Thursday. "When it's superintelligent, it'll be much better than people," he added.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in