Ask HN: What's your experience with GPT-Live been like? | Hacker News Reader

Ask HN: What's your experience with GPT-Live been like?

7 pointscaveman238 comments

For me it's been ambivalent:

It's conversational skills are an incredible leap from its predecessor: The acknowledgments it makes, and the natural flow it exhibits overall are awesome.

Today I began asking it something about Docker, and it said something along the lines of "I'm not too well versed with docker, but go on", which struck me as rather strange: Surely it "knows" more about docker than probably every greybeard out there, so howcome it shows this fake humility?

The other moment that stumped me was when I made a joke along the lines of "you're fired!" and it said "not funny", in a sort of nonchalant tone. I was like "what's that?", and it said "I said that was not funny." This seems like a far cry from its usual sycophancy and willingness to please human operator, no?

So this made me curious, what have some of your experiences with it been like? Anything noteworthy?

I use it to help with learning a language (French) and have been leaning into the CarPlay app. The model has done surprisingly well in conversations where it has to go back and forth across languages in the same response, and I like “talk to it while I’m driving” format a lot

I (people in general) don't talk to claude/chatGPT while brainstorming out aloud in the same way as a conversation with a person, in my experience. So I might pause a bit more to think about the request, and the new gpt-live feature tends to interrupt with 'human' sounding injections like "hmm", "huh" a bit too much for my liking. Also it's crutch is to start by paraphrasing what I just said to it which I'm sure is a grounding technique, but it's off putting, imagine having a real conversation like that, it can be good active listening but after every single thing I say it starts to feel a bit much.

GPT-Live isn't a particularly intelligent model. It needs to handoff for reasoning and more complex tasks.

You shouldn't expect it's responses to be as reliable as a frontier model. It will sometimes say odd things.

I've had a few extended discussions on a bit of maths (Zeno's Paradox) as well as Andy Burnham, the new British prime minister, and it's been extremely informative.

I also found it extremely useful at refreshing my understanding of arbitrary science concepts in playful manner, while doing chores and errands, making food etc.

One surprising thing is that my AirPods Pro could obviously deliver my voice to it quite clearly, even with the hoover going! I've never tested them under those conditions and assumed it wouldn't work.

Nice, yea I think the AirPods Pro got some of the best noise cancelling out there

never heard of it

Reply on news.ycombinator.com