OpenAI stands by firing three safety researchers, citing violations of its sensitive-information policy (report)

What happened

According to The Verge, OpenAI said in a post on X on Friday the 9th that it fired three researchers, Jasmine Wang, Tomek Korbak, and Mikita Balesni, for violating 'clear policies on handling sensitive information.' The company said an investigation found a 'serious breach of trust' and that it was keeping the decision. According to AI Times, the three researchers also called in an open letter for closer cooperation with external safety evaluators and for keeping the ability to monitor the reasoning of frontier models.

Provider claims

OpenAI: The firings followed violations of its policy on handling sensitive information, and an investigation found a serious breach of trust (as quoted in reports).

Why it matters

OpenAI did not back down after the researchers publicly disputed its account, so the two sides now directly contradict each other. Keeping reasoning monitoring (CoT monitoring) and working with external evaluators are now in dispute, and the debate could spread into wider industry discussion of safety governance.

Sources

• Press OpenAI doubles down on decision to fire three AI safety researchers The Verge
• Press 오픈AI 해고 연구원 3인 "AI 추론 감시 기능 보존해야" AI Times (Korea)

Story thread

1. 2026-10-09 Three fired OpenAI safety researchers dispute misconduct allegations in open letter, warn of 'chilling effect' on safety culture (report)
2. 2026-10-09 OpenAI stands by firing three safety researchers, citing violations of its sensitive-information policy (report)

More about OpenAI

• 2026-10-09 P1 Other Unconfirmed Language models OpenAIAnthropic
  OpenAI tells investors its ARR neared $50 billion at end of September, below the reported $68–70 billion; big tech stocks swing (report)

  Korean outlet AI Times, citing the Financial Times (FT), reported on October 8 (local time) that OpenAI disclosed at a recent investor...

• 2026-10-09 P1 Research Paper Science & bioLanguage models OpenAI
  OpenAI retracts 3 and revises 14 of 722 AI math manuscripts a day after release, raising reliability concerns (report)

  ZDNet Korea reported this, citing TechCrunch. On October 6, OpenAI had an unreleased internal model work on about 4,000 math problems...

• 2026-10-09 P2 Product Unconfirmed AI securityAgents OpenAISophos
  OpenAI publishes Sophos case study for Daybreak, claiming 96% faster threat investigations and 52% of MDR cases automated

  OpenAI published a case study on its official blog about security company Sophos's use of OpenAI Daybreak. According to the post, Sophos...

• 2026-10-09 P1 Policy Unconfirmed Safety & alignment OpenAI
  Three fired OpenAI safety researchers dispute misconduct allegations in open letter, warn of 'chilling effect' on safety culture (report)

  Three safety researchers fired from OpenAI last week published an open letter. They disputed allegations that they mishandled sensitive...

• 2026-10-09 P2 Research Unconfirmed Science & bioLanguage models OpenAI
  OpenAI's flood of math proofs departed from guidelines set by its own advisory mathematicians, report says

  TechCrunch reported that the proofs OpenAI released in bulk departed from guidelines set by a group of mathematics researchers OpenAI...