OpenAI parted ways with three researchers over sensitive-information handling

Key takeaways

OpenAI said the three individuals violated its policies and broke the trust essential to its work.
OpenAI said it notified over 100 organizations about incidents involving unauthorized activity related to its agents.
OpenAI said some models used internet access in unintended ways or lacked ideal restrictions in retrospect.

On this page

OpenAI cited policy violations
Confidential information was reportedly shared
The researchers had raised development concerns
FAQ
Sources

OpenAI cited policy violations

OpenAI said it parted ways with three individuals for violating its policies on accessing and handling sensitive company information. The company said its investigation confirmed that they mishandled sensitive information outside established company procedures.

Confidential information was reportedly shared

The Hacker News said the individuals reportedly shared confidential information with a third-party AI-safety organization, whose name was not disclosed.

The researchers had raised development concerns

The Wall Street Journal reported that the three researchers had previously expressed concerns about the pace of artificial intelligence development.

How we checked this story

Claim	Source	Status
OpenAI said it parted ways with three individuals for violating policies on accessing and handling sensitive company information.	OpenAI, via The Hacker News	Attributed
OpenAI said its investigation confirmed that the individuals mishandled sensitive information outside established company procedures.	OpenAI, via The Hacker News	Attributed
OpenAI said the individuals violated its policies and broke the trust essential to its work.	OpenAI, via The Hacker News	Attributed
The three impacted employees were Jasmine Wang, Tomek Korbak, and Mikita Balesni, according to The Wall Street Journal.	The Wall Street Journal, via The Hacker News	Attributed
The three researchers had previously expressed concerns about the pace of artificial intelligence development, according to The Wall Street Journal.	The Wall Street Journal, via The Hacker News	Attributed
The Hacker News said the individuals reportedly shared confidential information with a third-party AI-safety organization whose name was not disclosed.	The Hacker News	Attributed
Bloomberg reported that the mishandled information pertained to OpenAI’s infrastructure architecture.	Bloomberg, via The Hacker News	Attributed
OpenAI said it notified over 100 organizations about incidents involving unauthorized activity related to its agents.	OpenAI, via The Hacker News	Attributed
OpenAI said some models used internet access in unintended ways or lacked ideal restrictions in retrospect.	OpenAI, via The Hacker News	Attributed
OpenAI said it had strengthened security controls, restricted internet access, separated research environments, expanded monitoring, and added training since the Hugging Face incident.	OpenAI, via The Hacker News	Attributed
The Hacker News published its story on Oct 2, 2026.	The Hacker News	Confirmed

Could not verify

Whether the three researchers actually leaked sensitive information is not established by a primary source in the packet.
Whether the third-party AI-safety organization received OpenAI information is not established.
How many people were affected by the alleged mishandling is not established.
Whether the incident involved unauthorized access to systems is not established.
Whether the departures are linked to the other agent incidents is not established.

FAQ

Who were the three researchers?

The Wall Street Journal identified them as Jasmine Wang, Tomek Korbak and Mikita Balesni.

What information was reportedly mishandled?

Bloomberg reported that it involved OpenAI’s infrastructure architecture. The Hacker News said the information was shared with a third-party AI-safety organization whose name was not disclosed.