Skip to content
worth noting Business and investment verified update

Dismissed researchers at OpenAI warn of fear among employees in an open letter

confirmed by 2 independent sources updated 2 h ago

In an open letter, three dismissed researchers at OpenAI claim that the abrupt dismissals are causing fear among colleagues. They call for clear rules for working with external auditors. The company cites violations of rules for protecting sensitive information and denies retaliation for raising safety concerns.

In an open letter addressed to safety and advisory bodies at OpenAI, three dismissed researchers warn that their abrupt dismissals are causing fear and uncertainty among remaining employees about what conduct is permitted. They call for clear rules for working with external safety organizations, access to the company for auditors, and continued access to the reasoning processes of advanced models.

The Wall Street Journal identified the dismissed researchers as Jasmine Wang, Tomek Korbak and Mikita Balesni. According to Korbak, the company told him verbally that the reason was his communication with METR, which was part of his job responsibilities. Korbak believes the real reason was the safety concerns he raised internally. The researchers also deny being the source of a leak to The Information.

OpenAI says an internal investigation found violations of clear rules for handling sensitive information and a serious breach of trust. Its response does not specify the misconduct, and the company denies dismissing employees for raising safety concerns. According to the company, contracts with external safety auditors are being prepared. The reason for the dismissals therefore remains disputed between the company and the researchers.

What changed

Why it matters

The dispute raises the question of how employees can work with external safety auditors while protecting confidential information. According to the researchers, uncertainty about these rules may discourage colleagues from reporting problems; however, the sources do not establish an impact on model safety.

What was added since the original report

Verified updates

  1. New verified information

    In an open letter, the dismissed researchers warn of fear among remaining employees.; Tomek Korbak and Mikita Balesni were involved in investigating an incident involving the Hugging Face platform.; According to Korbak, OpenAI told him verbally that the reason for his dismissal was communication with METR.; Korbak claims that for months he had been raising concerns internally about the declining ability to monitor the reasoning of AI agents.; OpenAI rejects claims that it is losing the ability to oversee AI agents.

    • In an open letter, the dismissed researchers warn of fear among remaining employees.
    • Tomek Korbak and Mikita Balesni were involved in investigating an incident involving the Hugging Face platform.
    • According to Korbak, OpenAI told him verbally that the reason for his dismissal was communication with METR.
    • Korbak claims that for months he had been raising concerns internally about the declining ability to monitor the reasoning of AI agents.
    • OpenAI rejects claims that it is losing the ability to oversee AI agents.
  2. New verified information

    Wall Street Journal identifies the dismissed researchers as Jasmine Wang, Tomek Korbak and Mikita Balesni.; Tomek Korbak worked on the safety team.; Jasmine Wang and Mikita Balesni worked on aligning AI behavior with human intentions.; OpenAI has not confirmed the names of the dismissed researchers.; Korbak was the technical point of contact at OpenAI for METR and Redwood Research.

    • Wall Street Journal identifies the dismissed researchers as Jasmine Wang, Tomek Korbak and Mikita Balesni.
    • Tomek Korbak worked on the safety team.
    • Jasmine Wang and Mikita Balesni worked on aligning AI behavior with human intentions.
    • OpenAI has not confirmed the names of the dismissed researchers.
    • Korbak was the technical point of contact at OpenAI for METR and Redwood Research.

Relevant practical impact

What this means

01

For a business

For OpenAI, the dismissals have also become a dispute over the governance of safety research: the researchers are seeking clarification of the rules for external audits, while the company attributes the dismissals to violations of rules for handling sensitive information and is preparing contracts with auditors.

Risks and compliance
What to decide When working with external safety auditors, check whether the conditions for access and sharing sensitive information are explicitly defined.
More business impacts →
alignment AI safety Chain-of-Thought METR information protection OpenAI

Check the original

Event sources

confirmed by 2 independent sources · 2 publishers, 2 independent. We count feeds from the same owner only once.

3
TechCrunch AI independent context · first detected OpenAI cuts ties with 3 safety researchers, WSJ reports The Decoder (daily AI news) independent context Three firings and a fourth departure shake up OpenAI's safety team The Decoder (daily AI news) independent context OpenAI's safety crisis keeps getting worse and the company keeps making it worse