OpenAI Defends Firing Three Safety Researchers, Denies Retaliation

The dispute turns on whether internal safety advocacy and employment protections are compatible at a lab whose commercial pressures are intensifying. OpenAI's framing — a trust breach, not retaliation — is a reputational claim as much as a factual one, and it lands as safety staffing becomes a visible governance signal for the sector. What matters is whether the company substantiates the breach or leaves it as an assertion, and whether the departures shift how remaining safety staff weigh internal dissent.