That's my beef with most of the critical takes. Yeah, it has some warts, but an hour playing with the thing, for both broader chat interaction and writing code, make the value evident, and startling.
That's not because voice recognition doesnt work. It's been working quite well for 15 years now. Just ask the Deaf community.
The issue is even if the error rate is 1 in 100 and you scale that up to a billion users supporting thousands of use cases, you then need to maintain a large army of people to manually handle those errors.
Its a huge cost to even the largest firms. Very similar to what Facebook/Twitter/Youtube end up doing to handle content moderation. The faster systems scale the faster you get swamped with bugs you don't have bandwidth to fix. The consequences begin to multiply. Your basic Jurassic Park story plays out.
Comments
It takes about 5 minutes playing with ChatGPT to realize it has real time saving value, seems like author hasn't really tried to use it in their work.
While I agree there is a hype bubble forming, I disagree with the premise that it only produces garbage.
That's my beef with most of the critical takes. Yeah, it has some warts, but an hour playing with the thing, for both broader chat interaction and writing code, make the value evident, and startling.
The obvious thing is that we'll finally get voice assistants that can do more than pre-programmed tasks.
Right now I can basically ask my home assistant "what's the time/weather" and "play some music". Anything else returns "i don't understand".
I can see ChatGPT has huge potential here to give at least somewhat sane voice assistant responses.
That's not because voice recognition doesnt work. It's been working quite well for 15 years now. Just ask the Deaf community.
The issue is even if the error rate is 1 in 100 and you scale that up to a billion users supporting thousands of use cases, you then need to maintain a large army of people to manually handle those errors.
Its a huge cost to even the largest firms. Very similar to what Facebook/Twitter/Youtube end up doing to handle content moderation. The faster systems scale the faster you get swamped with bugs you don't have bandwidth to fix. The consequences begin to multiply. Your basic Jurassic Park story plays out.
I mean it returns "i don't understand" in terms of context.
"Who played bond in that film Goldensomething?"
"I don't understand."
ChatGPT on the other hand returns "you're thinking of Goldeneye, where Bond was played by Pierce Brosnan"