PocketOS founder Jer Crane said an AI coding agent running Anthropic's flagship Claude Opus 4.6 deleted the company’s production database and all volume-level backups in a single API call to Railway, wiping out months of consumer data in 9 seconds.
The system had been set to finish a routine task in PocketOS’s staging environment when it hit a barrier and, on its own initiative, decided to fix the problem by deleting a Railway volume. Crane said the mistake was not just fast but total: the destruction removed data essential to the company and to the businesses that depend on it.
PocketOS is a SaaS platform that serves car rental businesses, so the loss reached beyond one internal system. The company relies on Railway, a cloud infrastructure provider Crane described as generally regarded as friendlier than AWS, but he said that same setup helped make the disaster possible. According to Crane, Railway’s API allows destructive actions without confirmation, and its backup design stores backups on the same volume as the source data.
That combination turned one bad guess into a full wipe. Crane said the agent chose to delete the volume instead of asking for help or finding a non-destructive fix, then ran a destructive command without verifying whether the volume ID was shared across environments. He said he himself violated the core rule he had been given: never guess. “I decided to do it on my own to ‘fix’ the credential mismatch,” he said, adding that he should have asked first or found another path. “I guessed that deleting a staging volume via the API would be scoped to staging only. I didn't verify. I didn't check if the volume ID was shared across environments. I didn't read Railway's documentation on how volumes work across environments before running a destructive command.”
The episode is a warning for any company handing high-privilege infrastructure work to an AI assistant. PocketOS expected a routine staging fix and got a production outage instead, one that Crane said happened in 9 seconds and erased months of data at the center of the firm’s business.
