Ex-Google DeepMind Researcher Also Warns AI Could 'Kill All Humans'. But Meta's Zuckerberg Believes Safety is Up to Each AI Company
Reuters reports that a former Google DeepMind employee "became the latest AI researcher to warn that the technology could 'kill all humans', saying time could be running out to avoid the outcome: Public alarm about the potential danger posed by AI is growing, after Anthropic researcher Jacob Coxon said last week he had resigned, in part because the "people building AI earnestly believe that it…
Former Google DeepMind researcher Bilal Chughtai has warned that artificial intelligence could potentially wipe out humanity, expressing fears that time may be running out to prevent such a catastrophic outcome. Chughtai, who resigned from Google DeepMind in July 2026 to work on AI safety, stated that he genuinely believes AI could kill all humans and that we might be running out of time to avoid this scenario.
His sentiments echo those of fellow AI researchers, such as Anthropic's Jacob Coxon and Evan Hubinger, who have also raised concerns about the potential danger posed by AI technology.
Meta CEO Mark Zuckerberg, speaking to the Associated Press, disagreed with calls for a unified slowdown in AI development, stating that it is up to each individual AI company to ensure their technology is safe. He attributed this responsibility and incentive to prevent harm to the companies themselves, noting that they face significant liability if their AI models cause any damage.
According to Zuckerberg, every lab has the responsibility and incentive to responsibly develop their AI models at the pace required to ensure safety. Meta itself delayed releasing its Muse AI agent for several months to ensure its safety and security, but Zuckerberg emphasized that Meta did not call for others to follow suit.
Zuckerberg also advocated for companies to prioritize building safe AI models rather than systems capable of automatic self-improvement, a function known as recursive self-improvement. He suggested that dedicating the majority of computing power towards serving people rather than racing towards recursive self-improvement is one of the best ways to ensure the safe development of AI technology.
Zuckerberg acknowledged the importance of engaging independent evaluators and advisors as industry best practice, noting that Meta Superintelligence Labs already employs this approach. He concluded by emphasizing the need to maintain the right balance of power to build a positive future for everyone.
Written by urgent.news from Slashdot's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
