SAN FRANCISCO, Oct. 8, 2026 | Three AI safety researchers fired by OpenAI have written to the company's board urging it to preserve the ability to monitor how its models reason, according to Fox News' live coverage of a Wall Street Journal report. OpenAI says the firings were about confidential information, not safety concerns.
Jasmine Wang, Tomek Korbak and Mikita Balesni sent the letter to board members and safety committees. It has not been published in full, and the Journal has run only excerpts. "As an industry, we do not yet know how to safely develop and deploy models that we cannot monitor," the researchers wrote. They urged OpenAI to keep visibility into its models' reasoning, to work with independent auditors, and to avoid steps that reduce monitorability. They also said the dismissals are "chilling those who remain."
What the Letter Asks and Why It Matters
The central concern is chain-of-thought monitoring, which means reading a model's visible reasoning steps to catch unsafe behavior. According to Gizmodo's account of the letter, the researchers worry about OpenAI's flagship GPT-6 Astra. Gizmodo reports that the model's system card states that "Astra class models could evade our CoT monitors under adversarial conditions," and that its "recurrent depth" design cycles queries through the model internally, leaving less reasoning visible to monitor. Fox reports that the letter also asks OpenAI to work with independent auditors.
Korbak and Balesni were reportedly lead authors of an influential paper on chain-of-thought monitorability, according to commentary cited by Hugging News, which makes the letter a statement from people who helped define the field.
OpenAI's Account and the Open Questions
OpenAI announced Oct. 1 that it had "parted ways with three individuals for violating our policies on accessing and handling sensitive company information." Per the Journal's reporting, the three allegedly shared confidential material with an outside AI safety organization. Bloomberg reported that some of it related to the architecture of OpenAI's infrastructure, according to Fortune. OpenAI has not said what was shared or with whom.
When the Journal asked about the letter, an OpenAI representative said the firings "were not about raising safety concerns or speaking out." Fox reports that an internal OpenAI memo said the company "strongly agreed" with the researchers' recommendations and that "we do not terminate employees for raising concerns." The researchers dispute the reason for their dismissal, saying they never exceeded their professional duties.
The legal picture is unsettled. Fortune quotes Charlie Bullock of the Institute for Law & AI saying California whistleblower law generally covers disclosures to the government, law enforcement or internally, but not to private third parties, and that he withheld judgment on whether the firings were justified. Rep. Greg Casar said the firings looked like retaliation and that he would send OpenAI a transparency demand.
Frequently Asked Questions
Why did OpenAI fire the three researchers?
OpenAI says Jasmine Wang, Tomek Korbak and Mikita Balesni violated its policies on accessing and handling sensitive company information. The Wall Street Journal reports they allegedly shared confidential material with an outside AI safety group. The researchers dispute the reason, and OpenAI has not said what was shared.
What did the fired researchers ask OpenAI to do?
In a letter to OpenAI's board and safety committees, they asked the company to preserve its ability to monitor models' reasoning, work with independent auditors, and avoid steps that reduce monitorability. OpenAI says it "strongly agreed" with their recommendations in an internal memo.