Are you able to give a specific example of a problem it helped you solve? Especially one that you were at a complete blocking point, it provided some solution, and then you were able to continue to expand upon that solution.
I keep reading responses like yours, but I haven't seen any specific examples of problems being solved, so it all sounds very abstract. In my interactions with ChatGPT, it felt like just interacting with a search engine. There were zero continuations between questions and responses, and nearly 100% of the responses contained incorrect information.
Edit 1: As an example, I just asked ChatGPT to implement the Warren Abstract Machine for me. It gave me two different implementations, both with caveats that they are simple examples and not the whole thing, and neither implementation even type checked. It feels just like reading someone's homework where they copied off of someone else's work and had no idea what's going on. I don't see the point in this if it's just going to give me some high-level idea, which I already have, and an implementation that isn't remotely complete, much less one that doesn't even run. "Additionally, you'd need to ... handle backtracking". You don't say, ChatGPT?
Edit 2: I've kept on asking it to implement things I already know how to implement, even things that it probably has my code (being it was on GitHub) in its training data, and it keeps giving me code that doesn't even typecheck. ChatGPT is just doing what I've always imagined it doing: it's just doing a statistical merge. It has zero concept of anything I'm asking or it's saying.
Edit 3: I asked it something regarding WebGPU. It gave the typical, oh that's complex and "Here's a simplified example using GLFW and WebGPU through the F# WebGPU bindings provided by FableGL". But FableGL isn't a thing that exists, and even if it did, it wouldn't have anything to do with WebGPU which has nothing to do with OpenGL. And it imported Fable.Import.WebGPU and Fable.Import.GLFW, neither of which exist.
I mean, this is literally all smoke and mirrors. It boggles my mind when I hear people say they successfully use it every day. I haven't ever got it to tell me anything remotely correct.
I keep reading responses like yours, but I haven't seen any specific examples of problems being solved, so it all sounds very abstract. In my interactions with ChatGPT, it felt like just interacting with a search engine. There were zero continuations between questions and responses, and nearly 100% of the responses contained incorrect information.
Edit 1: As an example, I just asked ChatGPT to implement the Warren Abstract Machine for me. It gave me two different implementations, both with caveats that they are simple examples and not the whole thing, and neither implementation even type checked. It feels just like reading someone's homework where they copied off of someone else's work and had no idea what's going on. I don't see the point in this if it's just going to give me some high-level idea, which I already have, and an implementation that isn't remotely complete, much less one that doesn't even run. "Additionally, you'd need to ... handle backtracking". You don't say, ChatGPT?
Edit 2: I've kept on asking it to implement things I already know how to implement, even things that it probably has my code (being it was on GitHub) in its training data, and it keeps giving me code that doesn't even typecheck. ChatGPT is just doing what I've always imagined it doing: it's just doing a statistical merge. It has zero concept of anything I'm asking or it's saying.
Edit 3: I asked it something regarding WebGPU. It gave the typical, oh that's complex and "Here's a simplified example using GLFW and WebGPU through the F# WebGPU bindings provided by FableGL". But FableGL isn't a thing that exists, and even if it did, it wouldn't have anything to do with WebGPU which has nothing to do with OpenGL. And it imported Fable.Import.WebGPU and Fable.Import.GLFW, neither of which exist.
I mean, this is literally all smoke and mirrors. It boggles my mind when I hear people say they successfully use it every day. I haven't ever got it to tell me anything remotely correct.