4 ms·
They were afraid to release it because of unaligned output and hallucinations. ChatGPT showed that people could still get value out of something that wasn’t pe
by ec109685 3y ago
They were afraid to release it because of unaligned output and hallucinations.
ChatGPT showed that people could still get value out of something that wasn’t perfect.
E.g. they had this in their labs: https://www.theguardian.com/technology/2022/jun/12/google-engineer-ai-bot-sentient-blake-lemoine https://www.theguardian.com/technology/2022/jun/12/google-en... from July, 2022z
- halflings 3y agoAgree re:hallucinations/safety issues, that was likely one of the main blockers. And here's the sad part: they had this back in 2019... see this paper released in Jan 2020: https://blog.research.google/2020/01/towards-conversational-agent-that-can.html https://blog.research.google/2020/01/towards-conversational-...
- HarHarVeryFunny 3y agoLaMBDA was also briefly available for public testing, but then rapidly withdrawn due to unhinged responses. One advantage that OpenAI had over Google was having developed RLHF as a way to "align" the model's output to be more acceptable. Part of Google's dropping the ball at that time period (but catching up now with Gemini) may also have been just not knowing what to do with it. It certainly wasn't apparent pre-ChatGPT that there'd be any huge public demand for something like this, or that people would find so many uses for it in API form, and especially so with LaMBDA's behavioral issues.