Ranter
Join devRant
Do all the things like
++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatar
Sign Up
Pipeless API
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple API
Learn More
Comments
-
Lensflare2131416d> not what they wanted but definitely what they asked for
That’s hilarious 🤣
People started using AI without checking for correctness far too quickly.
Welp, that’s what they deserve I guess. -
D-4got10-01280415d> '(...) only do what's in the ticket description, the way it's detailed in the ticket description (...)'.
This sounds good solely on paper. In practice I see it failing more often than being right.
> 'IDK, it feels like the GPT is making people too optimistic, too lazy.'.
Lazy... definitely lazy, yes. -
There’s always been idiots and lazy people. They think AI is levelling them up but actually it’s exposing them.
In my very first job at a place that wrote bespoke software they were once asked by the client to add a large feature. It took months to add and we got paid for it.
Then when it was delivered another person at the client called to ask what all this stuff was and why we had done it.
We showed the spec etc signed off by them.
Turns out they didn’t want it.
Client thought we were joking when we they needed to pay us to undo all the work.
Think they got charged a months worth of time when all they did was revert the commits in source safe (this was a looong time ago)
Related Rants

hope this isnt a repeat. have a great weekend guys!!!!!
Reality
The client in the agreement asks us to be very strict about what we do, i.e. only do what's in the ticket description, the way it's detailed in the ticket description, otherwise we won't get paid.
okay..
Ticket description: "Perform database configuration analysis". We had a call where the client elaborated on what's expected from us (a full-blown DB configuration and security analysis). All good.
Ticket description: ChatGPT-generated, with all the unicode emoticons et al, bullet-pointing us what's expected. Basically, database auditing configuration analysis.
My guess is they never reviewed what ChatGPT wrote. And the service agreement is very clear and strict -- only do what's in the description. So yeah, 2 days worth of effort and I guess the result is not what they wanted, but definitely what they asked for (w/o them even bothering to read what they're asking...).
IDK, it feels like the GPT is making people too optimistic, too lazy.
rant
chatgpt
client
lazy
llm
ai