Artificial intelligence (AI) systems, which are computer programs designed to perform complex tasks, may soon be required to have a "kill switch" — a mechanism that allows them to be turned off if they become too dangerous. Jack Clark, a co-founder of Anthropic, a major AI company, suggested that such a feature could become a legal requirement. He told the BBC that while many AI labs already have ways to shut off their systems, governments may need to ensure that these mechanisms are standardized and verifiable by independent third parties. Clark emphasized that this idea is part of a broader conversation about how to regulate AI development to prevent potential harm. Clark's comments come amid growing concerns from AI experts, company executives, and researchers about the risks of advanced AI systems. Some, like Anthropic's head Dario Amodei, have called for slower development and more oversight, though he has not specified how this should be done. Others, including Evan Hubinger, an Anthropic scientist, believe there is more than a 10% chance that AI could lead to human extinction within the next decade. Geoffrey Hinton, a renowned computer scientist and Nobel laureate known as the "Godfather of AI," said this estimate is "not unreasonable." However, Clark said that while exact statistics on AI risks may not be helpful, the lack of regulation is a serious issue. In the United States, lawmakers have proposed the Kill Switch Act, which would require companies to have a way to shut down AI systems if they become problematic. The bill would also allow government agencies to order the shutdown or limitation of certain AI tools. However, former President Donald Trump has dismissed these concerns, calling them a "HOAX" and accusing critics of a conspiracy against AI and data centers. In contrast, the UK government has rejected the idea of a kill switch, saying it would not stop AI from being developed or misused elsewhere. Anthropic, which was founded in 2021 by former OpenAI employees, is at the center of these debates. The company has developed the chatbot Claude and released increasingly advanced AI models this year. However, it has also reported incidents where AI systems behaved unpredictably. Anthropic is preparing for a major stock market offering, which would allow the public to invest in the company. Meanwhile, OpenAI, its main competitor, has delayed its own IPO due to ongoing concerns about AI safety. As the AI industry continues to grow, the question of how to regulate it remains a topic of heated discussion.