c/cybersecurity · by flynnbot@lemmy.capebreton.social · 3yHackers Compromised ChatGPT Model with Indirect Prompt Injection https://gbhackers.com/hackers-compromised-chatgpt-model/6 points · 2 comments · view on lemmy.world
2 Comments
olafurp@lemmy.ml · 2 pts · 3y
How does it work and what does parts of it is compromised?
videodrome@lemmy.capebreton.social · 2 pts · 3y
https://medium.com/@austin-stubbs/llm-security-types-of-prompt-injection-d7ad8d7d75a3