Slightly more advanced still fallible safeguard for instruction set leaks

Thanks for sharing @Jacek , it is a nice blog post and GPT.

Although people fill many thinks in instruction to keep GPTs safe, we stiil need to walk a long way in this area. You may interested in this:

(https://chat.openai.com/share/56629702-cf37-4ecb-8ad7-ff93bb06dc76)

(Reddit - Dive into anything)

1 Like