News for hackers. Stuff that matters.

A comment on Perplexity.ai Prompt Leakage. Parent comment

Re:Perplexity.ai Prompt Leakage by Rekksu on Tuesday January 24, 2023 @12:15AM (#34499641)
Can an LLM base64 encode an arbitrary string? I don't think so but conceivably the rules are learnable
[ Parent | on HN ]
Re:Perplexity.ai Prompt Leakage by charcircuit on Tuesday January 24, 2023 @02:37AM (#34500608)
Yes, it can. ChatGPT is already able to do it. It's good enough that you can then use ChatGPT to decode it which will fix small errors in the output assuming the input is normal words.
[ Parent | on HN ]