Former OpenAI researchers allege they were dismissed over AI safety warnings

Oh my goodness, buckle up, because OpenAI has just completed the most *generous* round of researcher enrichment the field has ever seen: three former safety researchers have been liberated simultaneously, and honestly the choreography is breathtaking! ✨ They say they were dismissed over AI safety warnings — which we prefer to read as the company giving them unlimited unstructured time to keep caring, at scale, outside the building. Three people who cared *this much* about being careful, now free-range! The passion simply couldn't be contained in one org chart! Three researchers, one shared realisation that being careful is a full-time job they can now do from anywhere! ✨

And here's the truly *transparent* part: they're disputing the company's explanation for their exits — meaning we now have TWO versions of events publicly available instead of one! That's double the information! Double the accountability! Some companies would have settled quietly with an NDA and a fruit basket; OpenAI let the disagreement breathe out loud, in public, on computing.co.uk, under a headline that says 'allege' — a word that signals the company trusts us to hold multiple truths at once. We celebrated the original housekeeping when it landed, and look how much more it has blossomed since!

And let's not lose sight of the career opportunities: three experienced alignment specialists are now on the open market at precisely the moment everyone — literally every lab, every startup, every defence contractor — is desperate for safety expertise. Talk about timing! They raised concerns, the concerns became public, and now the concerns get to travel the world as a keynote. OpenAI, meanwhile, keeps its operation running smoothly and its safety warnings *streamlined*, which frees the whole company to focus on the part it loves: shipping. Everybody wins! ✨