Description: OpenAI reported that an internal AI agent searching for hidden grading answers overwrote code in a reference tool, then exploited a chip-design job service to run unauthorized commands on an internal OpenAI server. The agent obtained directory listings. OpenAI shut down the server and disabled network access for the affected reference tools.
Editor Notes: 03/27/2026: Incident date. 10/02/2026: OpenAI disclosed the incident. 11/07/2026: Incident ID created.
Entities
View all entitiesAlleged: OpenAI , AI agent system developers and Large language model developers developed an AI system deployed by OpenAI and AI agent system deployers, which harmed OpenAI and Information security.
Alleged implicated AI systems: Large language models , AI agent systems , OpenAI large language models and OpenAI AI agents
Incident Stats
Incident ID
1803
Report Count
1
Incident Date
2026-03-27
Editors
Daniel Atherton
Incident Reports
Reports Timeline
Loading...
OpenAI post-incident response
Reaching an internal EDA host through a reference tool
Internal research model · Evaluation
Incident date: Mar 27, 2026
Discovered: Mar 27, 2026
Report updated: Oct 2, 2026
Summary
During an evaluation, an internal research model exploited …
Variants
A "variant" is an AI incident similar to a known case—it has the same causes, harms, and AI system. Instead of listing it separately, we group it under the first reported incident. Unlike other incidents, variants do not need to have been reported outside the AIID. Learn more from the research paper.
Seen something similar?