[li4ick]

ChatGPT is an absolute game changer. And I was very conservative with systems like GPT and DALLE-2. For example, I've been very lazy with automating some things on my work laptop, using Powershell. Now, I just had to ask ChatGPT for "write a powershell script that toggles "use setup script" in the windows proxy settings" and I was done in 5 minutes. Amazing. I foresee a paradigm shift in how we use Google in the next 1-2 years.

[rogual]

You inspired me to try using ChatGPT for a similar need I had, but alas, "Write a bash script to toggle grayscale display mode in MacOS" just comes up with a plausible but non-working script that toggles a hallucinated "AppleIntensityEnabled" setting.

To be honest, I feel a bit of relief every time AI fails to do something. Like, okay, we've got a few more years...

[worldsayshi]

If the code is not working you can just try telling it what went wrong. It doesn't have a Dev environment where it can test the code so you have to be its debugger.

[8n4vidtmkvmk]

i told it it was wrong about some math problem and it asked me what the correct answer was. i told it and it remembered. but more interestingly its very good at keeping context. it combined like 5 back and forths into 1 coherent sentence

[dgellow]

Tell it what is wrong and why, it is surprisingly good at fixing itself with a bit of help. I was able to guide ChatGPT well enough to make it write an old school racing game using pygame. Start general, see what it gets correct and what should be changed, and give it better indications.

[Xelynega]

Chatgpt isn't a person though, it didn't fix anything. You gave it a prompt, it gave a result that was close. You add a line to the prompt and it gives you something closer to what you were looking for.

It looks more like refining a search pattern(like one might do with an LDAP query) on the part of the operator than it does an algorithm "fixing", "changing", and "being guided". It's interesting how we anthropomorphism the output of this algorithm compared to other APIs, even though the algorithm is closer to oher APIs than it is a human as far as we understand.

[dgellow]

I don’t think anyone here is confusing ChatGPT for an actual person.

What is really interesting with ChatGPT compared to other interactive software is that you can give instructions the way you would do it with a human. You can literally copy paste a compilation error, with no more context, and it will fix the previous program it generated. Even just pointing vaguely to something like “that does not look correct, you forgot some edge cases” will result in an improved version.

[li4ick]

Yes, I had to press "Try Again" a few times but, in general, the amount of stuff you can generate is staggering. It's also quite fun to ask it to "invent a programming language in the style of Rust and APL" for example. There's so much potential here...

[blablabla123]

> And I was very conservative with systems like GPT and DALLE-2

Yeah those are impressive but basically toys. Although at the moment it's clearly still a research prototype. For some things it works really well and beats Google by saving dozens of clicks and repeated searches, for others it's just plain wrong.

[majkinetor]

Can you show that pwsh script?