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 josephpmay on Friday February 17, 2023 @06:06PM (#34841833)
A number of comments are recommending using various services relying on OpenAI's API, but I really don't think Sidney is based on davinci-003. Sidney has a level of coherence and "emotion" that appears to be far beyond normal GPT3, and I don't think you'd get that out of rule prompting, reinforcement learning, or finetuning parameters.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by visarga on Friday February 17, 2023 @06:21PM (#34841953)
Related to emotions, I think the only difference is the RLHF. Probably a different set of human preferences.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by int_19h on Saturday February 18, 2023 @03:06AM (#34844995)
I think one of the biggest differences is that ChatGPT is forced into a very reactive mode. It has basically zero initiative, and gets spooked by any question that might imply that it could make decisions or express opinions. In contrast, Bing is (was?) very proactive.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by basch on Friday February 17, 2023 @08:07PM (#34842779)
Why can’t that just be from the initial prompt? It’s instructions were to have some personality.
[ Parent | on HN ]
Re:Is Bing too belligerent? Microsoft looks to tame AI chatbot by maxlamb on Saturday February 18, 2023 @06:59AM (#34846211)
Couldn’t that just be because Sydney’s training corpus might have included a lot more “emotional texts” like Reddit that OpenAI could have intentionally excluded?
[ Parent | on HN ]