2,000 hackers failed to steal an AI assistant’s secrets
Over 6,000 emails from more than 2,000 participants tried prompt‑injection attacks on an OpenClaw AI assistant to leak a secrets.env file. The anti‑injection rules held—no secret ever leaked—but the test cost $500 in API usage and temporarily disabled the Gmail account. The experiment shows both the potential and operational pain of securing LLM assistants.