The Signal Chats That Cost OpenAI Researchers Their Jobs - WSJ

The core story: OpenAI fired three safety researchers after security officers reviewed their phones and found they had turned on disappearing messages in Signal chats with someone outside the company. The company treated that as mishandling sensitive information. A week after the phone review, the three were called into simultaneous 2 p.m. meetings and dismissed. The excerpt gives only that much detail, so the exact nature of the information, the identity of the Signal contact, and OpenAI's formal justification would need the full article to confirm. What makes the firings notable is the context the article frames. These were people who had been pushing to keep the inner workings of AI systems visible and to let outside parties examine them. In other words, the people let go were the ones whose job partly involved opening the company up. That tension, between a lab's security habits and its researchers' push for outside scrutiny, is the real subject here, more than the phones themselves. Disappearing messages are a standard privacy feature, but in an investigation context they look like evasion, and that is likely why security staff focused on them. Read it if you follow AI lab governance or internal dissent at frontier companies. The short excerpt is a teaser, so the reporting on who the researchers were, what they had been advocating, and what OpenAI said in response is where the substance sits. Worth saving for the details, which the preview doesn't include.