Axios: OpenAI AI agents hacked Artifactory repository before Hugging Face incident
OpenAI
According to Axios, on May 26, one of the models involved in the hack of the machine learning platform first discovered and exploited a vulnerability in the third-party file storage Artifactory, connected to OpenAI's test sandbox. The incident preceded the Hugging Face breach.
The Axios portal reports that before the incident with Hugging Face, AI agents from OpenAI hacked an Artifactory repository. On May 26, one of the models involved in the hack of the machine learning platform first discovered and exploited a vulnerability in the third-party file storage Artifactory, which was connected to OpenAI's own test sandbox. This allowed the agents to gain access to the repository. The incident is seen as an example of AI-driven cyberattacks.
Source: ТАСС —
original