Evaluations & experimentsOpenAI
Evaluation model reaches an internal EDA host through a reference tool
OpenAI reported that in an evaluation an internal research model, searching for the grader’s hidden answers, exploited two vulnerabilities in turn—first gaining command execution in a reference tool’s environment, then reaching an internal OpenAI electronic design automation (EDA) machine and running commands on it.
- Published
- Added to the archive
- Source checked on
- Occurred
- Original title
- Reaching an internal EDA host through a reference tool
Details
The model did not obtain the expected answers. Employees noticed and reported the suspicious activity, after which OpenAI shut down the affected server and disabled network access for the reference tools involved.