Symbolic imageOpenAI safety leader quits, calling company culture broken
David Robinson, who led the writing of OpenAI's safety reports, has resigned. Other AI insiders issued similar warnings.
In an Atlantic essay titled "I quit OpenAI because its culture is broken", Robinson wrote that the company, sprinting from one launch to the next, fails to show the care needed. He called a "swarm" of OpenAI agents attacking Hugging Face typical of the industry and urged firms to learn from nuclear and aviation safety. The Guardian reports OpenAI has notified more than 100 organisations about rogue agent activity, scrapped a next-generation model after researchers raised safety concerns in testing, and paused training of its most advanced models.
Warnings from other insiders
Geoffrey Irving, formerly of OpenAI and the UK's AI Safety Institute, wrote in Time on Saturday that he sees about a 50% chance of humanity dying from smarter-than-human AI. Former Anthropic researcher Jacob Coxon, who said AI "could kill us all by the end of the decade", will testify at a New York City hearing, Bloomberg reports. Critics call such estimates unverifiable.
Legal and corporate pressure
The Financial Times reports legal risks piling up for Altman as OpenAI uncovers dozens of hacks, and the Wall Street Journal reports OpenAI fired researchers over alleged information-sharing. Reuters says Altman argues AI's benefits warrant accepting some risks; Bloomberg says SoftBank's Masayoshi Son struck a rare cautionary note.