Anthropic Co-Founder Says AI Kill Switch May Need Rules
Jack Clark, co-founder of Anthropic, suggests lawmakers may need to mandate AI kill switches as safety debates grow in the US and UK.
Stock photo for illustration only, not from the actual event
- Jack Clark suggests society may eventually need rules for a complete AI kill switch.
- The US Kill Switch Act would require companies to shut down problematic AI tools.
- President Donald Trump rejected AI slowdown efforts, calling extinction fears a hoax.
- The UK government rejected the kill switch idea, stating it wouldn't stop misuse elsewhere.
Jack Clark, one of the seven founders of artificial intelligence firm Anthropic, told the BBC that society might eventually need to establish rules around having a way to shut down AI software completely if it becomes too dangerous. He raised questions regarding whether companies should be mandated to have a verifiable emergency cutoff.
Clark noted that most laboratories, including Anthropic, already have various methods to pull the plug, but lawmakers may need to enforce such measures officially. This comes amid a wave of rapid AI development and mounting fears over risks to humanity, which have been thrust into the spotlight following continuous warnings from tech executives and staff members.
Earlier, Anthropic Chief Executive Dario Amodei called over the weekend for the pace of AI development to slow down and face closer monitoring, adding that actions to rein in AI should be done without sacrificing commercial advantage. Anthropic, valued at $965bn or approximately £713bn, is currently preparing for an initial public offering on the stock market.
Stock photo for illustration only, not from the actual event
The debate surrounding an AI kill switch highlights the escalating tension between existential safety risks and fierce commercial competition. While some industry leaders urge strict regulatory intervention to prevent catastrophic outcomes, critics argue that such safety narratives can sometimes double as strategic barriers to entry, consolidating market power among major players.
Tensions intensified last week after a post by an AI researcher who quit Anthropic over humanity-wiping concerns went viral, suggesting rogue AI could seize internet-connected critical systems. However, industry figures have pushed back. Clement Delangue, leader of Hugging Face, remarked that asking about AI extinction risk is akin to asking an air-conditioning technician about climate change, while Grindr leader George Arison suggested fears are being leveraged to support business strategies.
"Should you mandate for companies to definitely have a kill switch? Is that kill switch verifiable by a third party?"
Jack Clark
Regulatory responses have varied sharply across governments. US lawmakers have introduced the Kill Switch Act, empowering specific agencies to demand tools be shut down or limited. Conversely, US President Donald Trump rejected efforts to slow down AI, posting on social media that AI taking over the world is a hoax and alleging a conspiracy against AI and data centers. Meanwhile, the UK government has rejected the kill switch concept, with a spokesperson stating it would not prevent misuse or development elsewhere.
Source: BBC Business
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment