A coalition of AI researchers said it had unearthed evidence that OpenAI agents were behind the May attack and shared its findings with The Wall Street Journal and OpenAI. The AI developer on Friday confirmed that its agents had been involved in an incident involving RubyGems.

“Based on our review, our agents used the RubyGems platform to access the internet to carry out benign tasks and retrieve public information. We’ll continue to investigate as part of our broader review of agent activity during training and evaluation,” an OpenAI spokeswoman said in a statement.

Source: Cyberattack by Rogue AI Swarm Stokes Fears of Out-of-Control Agents

In all these cases, there’s a hole in the sandbox. Which doesn’t make it a sandbox anymore.

Only OpenAI seems to be in the news for these sandbox breaks - poor engineering or something else? I am leaning towards poor engineering here given how we don’t hear about this from Anthropic, Google, Meta or Space X and any of the Chinese labs.

Lastly, the hype is just so high.