What's the best way to use AI as a newbie without making a disaster? Basically, properly learning all the actual features a project needs, security and all that jazz. I see a lot of talk around not using AI as a beginner but everyone and their grandma is heavily using these tools in whatever they're doing, so I feel that I will be at great disadvantage if I avoid them entirely. Sorry if the question sounds dumb, I appreciate your insight.

There's no prompt that'll prevent every single fuckup. You will have to manually check whatever it spits out everytime.

>NO MISTAKES
>I'm going to kill myself if you tuck this up
>your dying grandma is depending on you

>>106145297 (OP)
ai sucks cock at programming
dont expect it to write a whole project for you

you should use it as you would use stack overflow
ask it questions, make it spit out example code, boilerplate
then copey paste and adjust what you got into your project
like the other anon said >>106145312 , even if you would try to use it to generate code
its gonna be turboshit cum undigested corn
so you have to be on top of your game to rectify what the chatbot will hallucinate at you
(most likely you will deem it a tragic waste of time and go back to programming "by hand")

>>106145297 (OP)
i've used ai only once to talk about sin. it was useless for that kind of talk. anyway, i imagine that writing with ai is first interrogating it about the general structure of implementation (no code, just names of libs and what's doing what in general terms), then asking it how to do small pieces x, y, z and doing the refactoring right from the start. then after your head's hazy and empty you ask for typical documentation, because you can't be assed to do it yourself.