Lee O'Brien-Riley
@leeobrienriley.bsky.social
Cyber and Technology Enthusiast & Ethical Hacking Fanboy!
Malicious AI prompts are now being sold online, enabling wannabe criminals to exploit ChatGPT for illegal activities.
Psst … wanna jailbreak ChatGPT? Inside look at evil prompts
Turns out it's pretty easy to make the model jump its own guardrails
www.google.com
January 25, 2024 at 6:55 PM
Malicious AI prompts are now being sold online, enabling wannabe criminals to exploit ChatGPT for illegal activities.