Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I've been using ChatGPT a lot. I'm no longer feeling much anxiety over the future of software engineering. GPT is a great tool. Still just a tool. It helps remind me how to do things I do rarely. To get code that exactly fits my requirements, I'd have to get so specific that I might as well just write the silly code myself. But for shell scripts, SQL, stuff like that, it's pretty decent.

Yesterday I asked it to give me a postgres SQL query to do something I don't often do, but I had a pretty good idea of what it would be. It confidently lied. I replied with "that is plausible but incorrect, I think you need to use X function instead" and it actually said "You're right, I'm sorry, that is the correct way to do X, here is an example". I laughed pretty hard at the casual apology followed instantly by the actually correct result. I'm in no danger of anthropomorphizing an LLM, but still.



Are you using 3.5 or 4? If 3.5, your experience is irrelevant since 4 is so much better.


I haven't noticed much of a difference. Can you give some examples where 4 has been much better?


"March 23 version"


4 costs money to use and the logo is black instead of light green. You’d know if you were using it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: