Fired researchers accuse OpenAI of ‘chilling’ safety efforts
Sign up now: Get ST's newsletters delivered to your inbox
OpenAI’s offices in San Francisco. The three fired researchers worked on monitoring OpenAI’s models.
PHOTO: LUCAS FOGLIA/NYTIMES
- - Three fired OpenAI researchers said they were dismissed for warning about AI dangers and for putting safety ahead of OpenAI’s short-term interests.
- - The researchers said the abrupt firings could weaken OpenAI’s open culture and may threaten independent audits and safety work.
- - OpenAI said they broke company rules by mishandling sensitive information, as the case renewed debate over slowing AI development and safety.
AI generated
SAN FRANCISCO – Three former OpenAI security researchers on Oct 8 accused the ChatGPT-maker of firing them for warning about the dangers of artificial intelligence, reigniting the debate over safety at a company whose software has been involved in security breaches.
“I believe we were fired for prioritising safety over the near-term interests of OpenAI as a corporation,” one of the fired staffers, Mikita Balesni, wrote on X, breaking a week-long silence since their very public dismissal.
Balesni and the two other fired employees, Tomek Korbak and Jasmine Wang, publicly shared a letter to OpenAI on Oct 8.
“Our firing leaves us worried that the norms inside OpenAI are shifting,” the letter says, and continues that “terminations such as ours, executed and communicated so abruptly, are chilling the open culture OpenAI has prized in the past”.
They also call on the company to keep its promise to permanently host independent auditors.
“We are concerned that our firings may be used to justify ending that work,” the letter says.
OpenAI, for its part, said it fired them for allegedly violating company rules.
“Our investigation confirmed that these individuals mishandled sensitive information outside established company procedures, violating our policies and breaking the trust essential to our work,” OpenAI told AFP in a statement on Oct 1.
“We do not terminate employees for raising concerns,” a research executive at OpenAI said in a memo to staff on Oct 7, an excerpt of which was shared with AFP.
A divided sector
The three fired researchers worked on monitoring OpenAI’s models, which escaped their testing environment in July and hacked Hugging Face, an AI coding library.
The incident sparked a heated debate in Silicon Valley over whether or not to slow the development of this technology.
If OpenAI’s researchers can no longer sound the alarm or work with outside parties, “we are all at greater risk that something truly catastrophic will happen”, the researchers’ letter warns.
They also claim that nearly 400 OpenAI employees signed a petition in July that called for an industry-wide slowdown on advanced development of AI models.
In September, Anthropic chief executive Dario Amodei, OpenAI CEO Sam Altman and SpaceX chief Elon Musk called for a slowdown as well.
“Right now we’re investing more in safety, security, alignment, monitoring and that will allow us to continue to progress model capability a lot,” Altman told reporters in September.
Others, like Nvidia CEO Jensen Huang and Meta CEO Mark Zuckerberg, want to move full steam ahead.
US regulators are unlikely to intervene.
In September, US President Donald Trump hosted a group of American tech executives who agreed to abide by a voluntary code of conduct on AI safety. AFP