Islamabad: OpenAI and three former safety researchers have taken their dispute into public view, trading conflicting claims over why the researchers were dismissed and whether their removal could discourage employees from raising concerns about artificial intelligence safety.
The confrontation involves Jasmine Wang, Tomek Korbak and Mikita Balesni, who published a letter warning that their sudden dismissals could create uncertainty among employees and weaken a workplace culture in which researchers had previously felt able to voice safety concerns openly.
Balesni also said in a post on X that he believed the researchers were fired for prioritising safety over the company’s immediate business interests.
OpenAI rejected that claim and said the dismissals followed what it described as a serious breach of trust.
The company did not publicly disclose the full nature of the alleged breach, but said the researchers were not terminated for raising safety concerns or challenging internal decisions.
OpenAI Rejects Claims Over Safety Concerns
OpenAI said it generally keeps individual employment matters private and does not consider public debate over personnel cases useful.
However, the company said it felt compelled to respond directly because of the claims made by the former researchers.
OpenAI maintained that employees are not dismissed for raising concerns and said it continues to encourage open, rigorous and critical discussion around product safety.
The company’s response has brought what had largely been an internal dispute into the public arena, placing renewed attention on the relationship between AI safety teams and the commercial pressures facing leading technology companies.
Former Researchers Warn of Chilling Effect
The three former researchers argued that the circumstances surrounding their removal could have a broader impact on OpenAI’s internal culture.
They warned that employees may become less willing to raise difficult questions if they believe doing so could put their positions at risk.
Their letter focused particularly on the importance of preserving an environment in which researchers can challenge decisions, question risk assessments and raise concerns about the safety implications of increasingly capable AI systems.
Balesni’s claim that the dismissals were linked to a preference for safety over near-term business interests further sharpened the dispute.
OpenAI strongly denied that account.
Core Disagreement Remains Unresolved
The disagreement now centres on two sharply different versions of the same events.
The former researchers argue that their dismissals risk sending a message that internal safety criticism is unwelcome.
OpenAI says the action had nothing to do with safety concerns and instead resulted from a breach of trust.
Because the company has not publicly detailed the alleged breach, the underlying reason for the dismissals remains disputed.
That lack of clarity has helped push the case beyond a routine employment matter and into a wider debate over how AI companies handle internal dissent, confidentiality and safety oversight.
AI Safety Culture Back in Focus
The public clash comes as the artificial intelligence industry faces growing scrutiny over how companies balance rapid product development with the need for safety research and internal accountability.
As AI systems become more powerful, researchers working on safety are expected to identify potential risks, challenge assumptions and raise concerns before technologies are deployed more widely.
At the same time, companies have strong incentives to protect confidential information, proprietary research and commercially sensitive material.
The OpenAI dispute highlights the tension between those two responsibilities.
For now, neither side has shifted its position.
The former researchers continue to argue that their dismissal could discourage open discussion around AI safety, while OpenAI insists the case was about trust and policy violations rather than criticism of the company’s safety practices.