What happened after 2k people tried to hack my AI assistant
A developer created a public AI assistant challenge to test security by inviting users to trick the bot into revealing sensitive files. The experiment resulted in thousands of emails, significant API costs, and a temporary suspension of the bot's email account due to fraud detection.
Why it matters
It highlights the significant security vulnerabilities and operational risks associated with deploying AI agents that have access to sensitive data and external communication tools.
I built hackmyclaw.com , where anyone could email Fiu, my OpenClaw assistant, and try to make it leak the contents of a secrets.env file.
The article is a first-person technical account of an experiment without political or ideological framing.
Get smarter about the news
Sign up free for a feed built around what you actually care about, Dive Deeper research on any story, and the full text of every article.
Create free accountAlready have an account? Sign in