Urgent advice from AI

A developer accidentally exposed an API token to an AI model, which responded with an urgent warning to revoke the credential immediately. The AI refused to proceed with the requested task until the security risk was addressed. The developer ignored the warning and did not change the compromised token.

Urgent advice from AI I once asked Composer 2 to add support for a third-party API in the code. I was so lazy that I just dropped the token into the chat and gave it the task. It replied in bold “This is urgent. You sent the token in the chat. Revoke it right now”. At least it wasn’t in all caps. It didn’t even try to do the task. No other model has ever talked to me in that urgent tone, like some cheap marketing gimmick. Of course, I didn’t bother changing the token.