What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Three OpenAI safety and alignment researchers, Jasmine Wang, Tomek Korbak, and Mikita Balesni, say they were dismissed the week before October 8, 2026, and that OpenAI’s explanation does not match what they did. OpenAI says an investigation found a “pattern of misconduct” involving mishandled research information, and an internal memo says the decisions were not retaliation for raising safety concerns. The researchers warn that abrupt terminations and unclear rules could make colleagues afraid to speak up and to work with outside safety evaluators. The available reporting and the researchers’ open letter do not settle who is right, so each side’s claims are set out below as attributed positions.
What happened
TechCrunch reporter Rebecca Bellan reported at 1:04 p.m. PDT on October 8, 2026, that the three researchers had published an open letter addressed to OpenAI’s Safety and Security Committee, its Safety Advisory Group, and its Mission Advisory Council. The letter identifies the authors as the three safety and alignment employees dismissed the previous week. The full letter is hosted on Balesni’s personal site, and it is the primary source for the researchers’ account: the open letter (PDF). The news report is at TechCrunch’s report.
As an Amazon Associate I earn from qualifying purchases.
OpenAI’s stated position
As reported by TechCrunch, OpenAI told the outlet that an investigation revealed a “pattern of misconduct” involving clear violations of policies on mishandling research information. According to that account, the conduct went beyond sharing information with an outside AI evaluation group. OpenAI did not, in its response to TechCrunch, identify the specific policies it said were violated, and it did not describe the circumstances of each dismissal.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Repair Windows errors before they cause bigger problemsFix Now →A memo attributed to an OpenAI research leader praised the three researchers’ contributions and rejected the retaliation claim. TechCrunch quoted it as saying: “I want to be very clear that these decisions were not about raising safety concerns or speaking out. We have always encouraged that and always will. We do not terminate employees for raising concerns.”
#1 Best Overall
TechCrunch also reports that OpenAI had not formally responded to the open letter at the time of publication.
The researchers’ account
The three researchers say they acted within the working norms and mandates they understood at the time. They deny being the source of a leak to The Information about less-monitorable architectures, and they say they did not engage external parties outside their job mandates. Their letter describes the investigation into a Hugging Face incident as unprecedented, with internal policies still being developed while it was underway. These are the authors’ assertions, not the findings of an independent review.
The letter gives a separate account for each person.
Free tools Windows power users keep installed
One-click scans. No signup required.
Tomek Korbak
The letter says Korbak was the technical contact for METR during the Hugging Face incident investigation. The authors argue that careful communication with external counterparts mattered for building trust while procedures were still taking shape.
Rank #3
Mikita Balesni
According to the letter, Balesni worked on cross-company efforts to preserve the monitorability of AI models, communicated with board members and executives, checked with his reporting line, and removed sensitive details before sharing materials.
Jasmine Wang
The letter says Wang had been delegated access to executive email for recruiting purposes and asked IT to remove that access once it was no longer needed. According to the letter, IT did not remove it. The authors write that the inboxes looked identical in her mail application, and that she accidentally opened a sensitive email addressed to an executive within minutes. She reported this and again asked IT to remove the access.
TechCrunch separately reports Wang’s account that OpenAI told her she was fired for accessing an executive’s email. The email-access explanation and the outside-collaboration questions are different issues, and the reporting does not show that one caused the other.
Why the researchers say it could chill safety work
The central argument of the letter is that abrupt dismissals and unclear rules make current and former colleagues unsure where they stand. The authors say this could make people less willing to speak openly and to continue work they regarded as central to OpenAI’s safety culture, and that it could weaken accountability from outside safety groups. The letter presents this as a warning about a likely cultural effect. Neither the letter nor the news report measures that effect.
Best Value
The letter’s three recommendations are:
- Honor public commitments to embed independent third-party safety auditors and continue collaboration with the wider safety ecosystem.
- Preserve the monitorability of frontier models. The authors say monitorability is degrading and argue against developments that further reduce it, since safety depends on monitoring.
- Keep open dialogue between internal safety researchers and external safety organizations, with clear written procedures for permitted collaboration.
The letter also refers to a public commitment by Sam Altman on September 12 to give independent evaluators ongoing, employee-like access. This article has not verified the terms or current status of that commitment.
The letter’s own framing is the clearest summary of the authors’ concern: “The freedom to do so without fear, and to have well-defined internal procedures that enable this work, is itself an essential safety mechanism.” They also write that “Terminations such as ours, executed and communicated so abruptly, are chilling the open culture OpenAI has prized in the past.”
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Comparing the two accounts
| Question | OpenAI, as reported by TechCrunch | The three researchers, per their open letter |
|---|---|---|
| What conduct occurred? | An investigation found a pattern of misconduct, including clear violations of policies on mishandling research information, beyond sharing information with an outside AI evaluation group. | Work-related collaboration with external parties, the handling of a Hugging Face incident investigation, and an access problem with executive email. They deny being the source of the leak to The Information and deny engaging outside parties beyond their mandates. |
| What rules applied? | Policies on mishandling research information. The specific policies were not named in OpenAI’s response to TechCrunch. | Working norms and mandates they understood at the time. They say internal procedures were still being developed during the investigation. |
| Were the dismissals retaliation? | No. The internal memo says the decisions were not about raising safety concerns or speaking out. | The letter does not claim the dismissals were formally retaliation. It argues that abrupt terminations and unclear rules are chilling open debate. |
| Effect on safety work | Not addressed in the reported response. | Fear and unclear rules could hinder safety work and weaken third-party accountability. No measured outcome is provided. |
Keep three threads separate when reading either account: the alleged leak, the Hugging Face investigation, and Wang’s email access. The reporting and the letter discuss them together, but they are distinct allegations.
What is not yet established
- The specific policies. The available sources do not identify which OpenAI policies were allegedly violated.
- The investigation. Neither the full evidence nor the process behind OpenAI’s investigation has been published.
- Each dismissal. The circumstances of each termination are described only from the researchers’ side, except for Wang’s report of what she was told.
- The chilling effect. No independent source measures whether employees have become less willing to speak up or to work with outside evaluators.
- The internal norms. No independent review confirms the researchers’ description of the procedures that applied at the time.
Until OpenAI responds to the letter or publishes more detail, the dispute rests on competing attributed accounts. Readers should treat the researchers’ claims, OpenAI’s claims, and the predicted chilling effect as separate questions with different levels of support.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
