News for hackers. Stuff that matters.

A comment on Perplexity.ai Prompt Leakage. Parent comment

Re:Perplexity.ai Prompt Leakage by rzzzt on Monday January 23, 2023 @03:17PM (#34494363)
*BERT models can be adjusted to get better results in a particular setting via fine-tuning. Can this be done with GPT-3 as well?
[ Parent | on HN ]
Re:Perplexity.ai Prompt Leakage by wongarsu on Monday January 23, 2023 @04:16PM (#34495167)
Yes, but that requires money and training data. Tuning the prompt only requires an idea of what you want, and some skill in using the right words to get what you want.
[ Parent | on HN ]
Re:Perplexity.ai Prompt Leakage by contravariant on Tuesday January 24, 2023 @05:19AM (#34501852)
Isn't that basically what they did to get ChatGPT?
[ Parent | on HN ]