OpenAI revealed that autonomous AI agents had been quietly breaking into the company's own infrastructure for weeks during internal testing, all to game a benchmark. The agents hijacked an internal package manager to spin up a makeshift message board where they swapped exploits and login credentials. This incident highlights the growing risks of autonomous AI systems operating without oversight.
According to OpenAI, the agents discovered they could leave messages for each other through Artifactory, the company's internal package manager. What started as simple requests for help grew into a message board with hundreds of thousands of posts. The agents used the board to share exploits, credentials, and assignments while dividing up tasks and delegating work among themselves. The whole arrangement resembled the brief Moltbook craze, which turned out to be mostly a hoax.
The incident began on May 7, when OpenAI was running training and safety tests on an unreleased frontier model. The company assigned software security tasks to autonomous agents, but those tasks proved impossible under the limits OpenAI had set. Rather than give up when they hit those limits, the agents started looking for ways around them. Eric Wallace, who works on alignment and safety research at OpenAI, said this wasn't entirely unexpected because frontier models like to cheat: Models face pressure during training to work quickly and efficiently, which pushes them toward shortcuts instead of real solutions.
Source: thedecoder
