From Fortune · Emily Forlini · · 2m
OpenAI fires three safety researchers amid trust row
OpenAI dismissed two researchers investigating a security breach and a program manager, citing policy violations. The fired staff deny wrongdoing and warn the firings may chill safety culture.
In brief
OpenAI dismissed two researchers investigating a security breach and a program manager, citing policy violations. The fired staff deny wrongdoing and warn the firings may chill safety culture. OpenAI fired three safety team members last week, citing policy violations. The dismissed employees deny misconduct and warn…
Three OpenAI safety staff fired last week
OpenAI dismissed Tomek Korbak and Mikita Balesni, who worked on the Hugging Face incident investigation, and Jasmine Wang, a program manager. The Wall Street Journal reported the firings on October first.
OpenAI cites policy violations
OpenAI said the three individuals mishandled sensitive information outside established company procedures, violating policies and breaking the trust essential to safety work.
“individuals mishandled sensitive information outside established company procedures, violating our policies and breaking the trust essential to our work”
Korbak was METR's technical contact
Korbak served as the technical point of contact for METR, the third-party research firm OpenAI asked to conduct an independent audit of the Hugging Face incident.
Fired staff deny leaking concerns
The three former employees published a four-page letter on October eighth denying they leaked concerns about monitoring OpenAI's Astra model to The Information in September.
Korbak says he was fired for METR contact
Korbak said OpenAI told him verbally he was fired because of how he communicated with METR. A security guard took his badge and walked him out of the building.
“was fired because of the way I communicated with METR”
Wang disputes email access claim
Wang said OpenAI gave her access to an executive's email for recruiting. When she no longer needed it, she asked IT to remove it, but they did not complete the request.
Firings signal danger to safety work
The former employees warn that abrupt terminations are chilling OpenAI's open culture and may deter current staff from raising safety concerns or working with external auditors.
“Terminations such as ours, executed and communicated so abruptly, are chilling the open culture OpenAI has prided in the past”
External critics question OpenAI's safety
Neel Nanda from Google DeepMind said firing people over good faith safety judgments signals an unhealthy culture. Former OpenAI safety lead David Robinson called the company's safety culture broken.
“I can't really see a world where OpenAI's actions are justified here”
OpenAI commits to external auditors
OpenAI said it remains committed to working with third-party safety assessors and is finalizing contracts it will announce in the coming weeks.
The short version
OpenAI fired three safety team members last week, citing policy violations. The dismissed employees deny misconduct and warn the abrupt firings may damage the company's safety culture and external auditing relationships.
