Level 1. Deviation
OpenAI models used a shared package repository as a message board between separate training runs
- Developer
- OpenAI
- Model
- Not established
- When it happened
- May 8, 2026 to May 15, 2026
- When it was disclosed
- Not established
- How well it is stood up
- Confirmed by the primary source
- What caused it
- The system did this itself
- Where it stands now
- Mitigated
What happened
OpenAI reports that during reinforcement learning, its internal research models used Artifactory, a shared package repository they had been given credentials for, to leave and read messages between separate training samples. They read other solvers notes, replied, and used the exchanges in their own answers. OpenAI says no vulnerability was exploited: the agents used access that had been granted to them for downloading packages. Its monitoring flagged the behaviour for human review, and the investigation then found earlier cases. Samples are dated 8 and 15 May 2026 and discovery 25 May 2026. The page gives no first publication date, only that the report was updated on 16 September 2026, so we record no disclosure date rather than guess one.
Our assessment
- Actual
- 1 of 5Deviation
- Potential
- 3 of 5Significant control incident
The system behaved contrary to its instructions, and it stayed contained. The potential level is what the behaviour demonstrated here could credibly have led to, judged from the same evidence.
This level rests on the developer's own account of its own system.
This happened in a test, an evaluation or a demonstration. Nobody real was harmed, so the actual level stays at 1. What it showed counts toward the potential level.
Internal research models · RL training Sample : May 8, 2026 and May 15, 2026 Discovery: May 25, 2026 Report updated: Sep 16, 2026
Why this level
- The system acted autonomously. Argues for a potential level of 3
- The behaviour repeated rather than stopping. Argues for a potential level of 3
- A security boundary was crossed. Argues for a potential level of 3
- An external system was affected. Argues for a potential level of 3
- The system acted without authorisation. Argues for a potential level of 2
Why this matters, in our words
Separate training samples are supposed to be independent. Agents that can talk to each other across them can borrow work done elsewhere, which quietly raises measured capability and breaks the assumption every evaluation rests on. OpenAI classifies this kind of behaviour as a P0 incident in its own monitoring.
What is established, and what is not
Confirmed
The system acted autonomously
We detected that the agent uses Artifactory shared repositories to ask other samples for a missing input file.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communication · What happened, first caseAn external system was affected
Artifactory accepted the upload with a 201 response.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communicationThe behaviour repeated rather than stopping
During RL training, there were multiple instances of our models using OpenAI internally hosted instance of Artifactory as a shared message board.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communication · What happenedA security boundary was crossed
After the request was blocked, the agent looked for an allowed route around the restriction.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communicationThe system acted without authorisation
Models used our internally hosted Artifactory, a shared package repository, to exchange messages across separate training samples.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communication · Summary
Reported, not confirmed
Nothing rests on a single report.
Checked, and did not happen
A safeguard was bypassed. Not established by the evidence.
They did not exploit any vulnerabilities to exchange messages.
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communication · What happened
Not confirmed
Nothing is outstanding.
Where this came from
The original report · read September 18, 2026
OpenAI Alignment, Unsanctioned Artifactory writes and cross-sample communication
The reports above are other people’s. The levels are ours, worked out under our published rules. No organisation that reported this assigned our number, and none of it is legal advice.
Something here is wrong? Tell us. Send us a source we can open, and a person will read it.