News for hackers. Stuff that matters.

A comment on Is Bing too belligerent? Microsoft looks to tame AI chatbot. Parent comment

Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by sp332 on Friday February 17, 2023 @05:22PM (#34841430)
You can sign up to use the underlying language models, like davinci-003. https://platform.openai.com/docs/models/overview But you will have to come up with your own prompts. You can even pay to fine-tune it for your own tasks.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by basch on Friday February 17, 2023 @05:23PM (#34841439)
You can input the Sydney prompt, no?
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by eightysixfour on Friday February 17, 2023 @05:27PM (#34841483)
It is very unlikely that the underlying Sydney model is Avi able, you can approximate it, but I have never had davinci create nearly as much personality as “Sydney” has.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by LesZedCB on Friday February 17, 2023 @05:27PM (#34841477)
ChatGPT and Syndey were extensively trained with chat specific training via RLHF (realtime learning from human feedback), which is missing from GPT-3.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by warkdarrior on Friday February 17, 2023 @05:44PM (#34841658)
RLHF = _reinforcement_ learning from human feedback
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by visarga on Friday February 17, 2023 @06:26PM (#34842005)
Haha, someone was asking today on Twitter

> I can't find a mention of "GPT" or "General Purpose Transformer"

:)

[ Parent | on HN ]