Silicon Valley's latest doomsday talk isn't coming from outsiders—it's coming from the people building the machines. Three Anthropic staffers have gone public with warnings that advanced AI systems could wipe out humanity within years, not centuries, reports Axios. Researcher Jacob Coxon, who resigned from Anthropic this week, wrote that many AI builders "earnestly believe" their work could end human life "by the end of the decade," adding that executives often soften their language in public while privately expressing fear.
"No other human activity poses this level of danger," Coxon wrote. He adds to the Wall Street Journal: "We're on track for a lot of the most aggressive of these scenarios where by the end of next year things could be out of control already." Evan Hubinger, who leads alignment science at Anthropic, backed Coxon up, saying he sees more than a 10% chance that AI could kill "all humans" in the next decade and that the company lacks a clear plan to make superintelligent systems safe. Scalable-oversight lead Samuel Marks, meanwhile, noted that said concerns tend to grow with seniority inside AI labs.
Their warnings echo recent comments from OpenAI leaders, including CEO Sam Altman, who say AI is racing into a frightening phase that may be impossible to manage, per Axios. Skeptics contend that such alarm boosts valuations and cements incumbents' power via regulation beneficial to them, but reporters who've spoken with insiders for months say the unease appears genuine, and increasingly urgent for regulators.
,
uBlock 
,
and
Ghostery 
)
