interesting…. could the RLHF tease more of that out of them??
@willknight
-
Understanding AI Model Anthropomorphization and Training Data
By
–
it seems like one of the inherently confusing things about these models. I know that they've learned to talk about their feelings and desires because that's in the training data, but I think it contributes to the misunderstanding.
-
Google Bard Easter Eggs Borrowed From Assistant Voice Helper
By
–
I wonder what the folks behind Google Assistant make of Bard? It claims to have some easter eggs, but they seem to be borrowed from Google's voice helper.
-
Future Cars Will Monitor Driver Behavior and Safety
By
–
In the future, your car will be watching you. Nice story by @AarianMarshall on the future of driver-assist. https://
wired.com/story/cars-tha
t-watch-their-drivers-could-re-teach-the-world-to-drive/
… -
Daily human sacrifice at ChatGPT altar
By
–
Looks like a regular day of human sacrifice at the alter of chatgpt
-
ChatGPT hype and AGI speculation: corporate responsibility needed
By
–
Seems like hype around ChatGPT is distorting things (wild speculation about what it will do and even talk of AGI). Microsoft and OpenAI could do more to keep that in check.
-
Autonomous Driving vs Language Models: Maturity Comparison
By
–
Autonomous driving perhaps? Probably not as new and unproven as these language models though.
-
Noam Shazeer Creates Bespoke Model Based on LaMDA
By
–
No, it's a bespoke model. The founder is Noam Shazeer, one of the creators of LaMDA.
-
Microsoft Google LLM Deployment Risks and Reliability Concerns
By
–
It's a bit strange to watch Microsoft and Google rush to deploy generative AI while ever more problems with these systems emerge. LLMs have amazing potential but they are born fabricators (see bing query). Would be good to see more focus on making these tools less chaotic.
-
How AI Models Conform to Leading Questions in Responses
By
–
I don't think you're being misleading at all. What I find interesting though, is how much the model seems to want to find answers that fit with leading parts of the question. If you'd asked "what role microsoft didn't play", presumably it would've run with that.