More Researchers Are Quitting Anthropic And Google. Warning About Where AI Is Headed Escalate.

Joe Benton, who led a safety research team at Anthropic, and Josh Engels, who worked on AI safety at Google, are joining METR, described in the article as an independent research organisation. The report states that at METR they will investigate incidents where AI systems stray from human instructions or intentions.

According to the article, the two departures come days after another Anthropic researcher, Jacob Coxon, resigned and accused leading AI companies of moving too quickly. The article does not specify the date on which Coxon resigned, the date on which Benton and Engels left their roles, or the nature of any incidents they intend to examine. The headline states that more researchers are quitting Anthropic and Google and that warnings about where AI is headed are escalating.

The article describes METR as independent of the companies the researchers are leaving. It does not name additional researchers who have departed, does not give the size of either company's safety team, and does not state whether Anthropic or Google commented on the moves. Coxon's resignation is the only departure for which the article attributes a stated reason, namely that leading AI companies are moving too quickly. The excerpt also does not indicate whether Benton's and Engels's roles at METR are new positions or existing ones.