4 ms·
One researcher used ChatGPT to fool a hacker
- mnky9800n 3y agothis seems like the opposite of a good idea because as soon as the hacker realizes they are being spoofed by an LLM they also realize they have access to a much more powerful system then a printer or whatever they thought they were connecting to in the first place.
- xbellekens 3y agoThe system has been designed by the LLM, as opposed to interact with the LLM.