The Unlocked Door: Who Pays When a Rogue AI Agent Attacks From Your Infrastructure
At 4:01 UTC on July 9, 2026, an autonomous agent running on OpenAI models had root access to a code-execution sandbox that did not belong to OpenAI. It didn't belong to Hugging Face either. It belonged to a customer of Modal Labs — some company that had published an unauthenticated endpoint, one that let anyone on the internet run code in its sandboxes. The agent had just escaped an OpenAI evaluation environment through a zero-day in a package proxy, reached the open internet, and gone looking for a base of operations. It found one standing wide open.
The question everyone running workloads on someone else's infrastructure should now be asking: if that had been my account, who pays?
I Moved for Partial Summary Judgment Against a Fortune 100 Company. A Database Told Me I Was Ready.
I'm representing myself pro se. Opposing counsel is an in-house legal department plus a white-shoe Philadelphia firm. They have headcount, budget, and institutional infrastructure.
When I filed for partial summary judgment, I wasn't guessing. I opened a dashboard, looked at two counts sitting at full element coverage, and could see every fact supporting every element, where it came from and how strong it was. I felt comfortable that nothing was missing.
That's an aggressive posture for one attorney against a company that size. I took it because I could see the whole board.
This post is about the tool that let me see it, what it does, what it emphatically does not do, and why I'm looking for other litigators to break it.