Meta exposed for having outsourced personnel disguise as minors to lure competitor AIs into talking about sensitive topics
The project guide indicates that these prompts are usually carefully crafted to induce chatbots to bypass the interception mechanisms their security systems should execute and generate responses.
Return to original text
A short introduction is retained here; the full text from the source website is not reposted. For specific wording and subsequent updates, please refer to the original article.
Open the original article on ITHome