OpenAI Fires Researchers Who Raised Safety Concerns, Claiming a Security Breach

Fri Oct 09 2026 12:21:03 GMT+0000 (Coordinated Universal Time)

OpenAI fired researchers for raising safety concerns, sparking a debate over AI governance.

Three former OpenAI staff allege they were let go for prioritising AI safety over corporate interests. A joint letter on X claims the dismissals were punitive for whistle‑blowing, while the firm insists data safeguards were breached. The clash highlights rising tensions around AI governance and the need for safer oversight. Experts warn that such incidents could chill internal scrutiny of advanced AI systems and erode public trust in the industry. The broader debate raises questions about how to balance rapid innovation with ethical safeguards and transparent accountability.

OpenAI fires researchers who raised safety concerns, claim dismissal over security breach

Three former OpenAI researchers say they were sacked for championing AI safety, rather than for mishandling data, following an open letter they posted on X.

Allegations of safety‑first dismissal

"I believe we were fired for prioritising safety over the near‑term interests of OpenAI," posted Mikita Balesni on X, joined by colleagues Tomek Korbak and Jasmine Wang who had all worked on safety alignment.

At the time, an OpenAI spokesperson said an investigation found the trio had "mishandled sensitive information outside established company procedures." The firm denied any link to safety protests, claiming the case was about breaches of trust.

OpenAI’s official statement

An OpenAI research leaders’ note climbed the company’s X account on Friday: "We want to be very clear that these decisions were not about raising safety concerns or speaking out. Our internal investigation uncovered a significant breach of trust beyond what was outlined in the letter." The note also confirmed the firm was finalising contracts with third‑party safety assessors.

Reactions and wider context

Wang warned that if staff "take a stand now against this kind of manoeuvre" the situation could repeat with other employees. The incident follows a broader debate over AI safety and the danger of advanced systems that could cross‑hinder human values.

The letter pointed out the firings had left many OpenAI staff "afraid to speak and operate in ways that, until last week, were integral to their work" and highlighted the loss of monitoring tools that flag misbehaviour in AI agents.

With OpenAI’s ChatGPT and other models under scrutiny for safety and policy compliance, the firing dispute adds to calls for clearer governance and whistle‑blower protections in the AI industry.