The LLMs are just somewhere between an averaging and a lossy compression of everything on GitHub. There's nothing about the current paradigm of "AI" that is going to somehow do better than just rehashing that training set but with the inclusion of various classes of errors.
I think it's better to view it as spicy search rather than any form of intelligence.
21
MrScottyTay - 1w
Yeah that's mostly what i use it for. A way to search for things that i can't describe well enough for a traditional search engine so I can find out at a glance what it's name is and if it's valid for my situation. If it is then I go look up documentation. Any time I've stayed in the LLM past that I eventually go down a rabbit hole of wrong ideas that aren't always obvious until you get a bit too deep and you've wasted an hour with an incorrect solution.
8
Corkyskog @sh.itjust.works - 7day
The only thing I have found it useful for is book recommendations. I like this book and that book, what other types of books are like these?
3
potatopotato @sh.itjust.works - 7day
Yeah, I do believe it's a good tool for search, just with the caveat that if it can't find an answer it makes one up or otherwise kinda just fills in little missing details with noise.
2
Serinus @lemmy.world - 1w
If you treat it as spicy search it works pretty great though.
3
doo @sh.itjust.works - 6day
As I'm slowly evolving my own flavour of spec driven development, I'm starting to think about the generated code as a secondary artefact where main quality criteria is that it's doing what it needs to and it's covered with tests.
I guess my current analogy is that I don't care about how readable or dry is the assembly code generated by compiler.
I have the specifications and the working code with tests. I can always regenerate it if I need to.
But. I still read the produced code, steer the design and correct the obvious blunders. No vibes.
-1
utopiah - 6day
main quality criteria is that it’s doing what it needs to and it’s covered with tests.
Might want to read on TDD, it's been around since last the last millennium (OK 1999 according to Wikipedia, point is, it's not new).
3
monkeyslikebananas2 @lemmy.world - 6day
I know a lot of people hate but this AI stuff still isn’t great but it will get better. Each generation of programming languages adds syntax and convenience. AI code will likely get to the point where it is just a higher level language. The only benefit I’m seeing is that if used very carefully I can make more complex projects with fewer team members. And where there was zero documentation there’s at least SOME documentation.
0
Arthur Besse - 5day
AI code will likely get to the point where it is just a higher level language
cypherpunks in technology
AI Is still making code worse: A new CMU study confirms
https://blog.robbowley.net/2025/12/04/ai-is-still-making-code-worse-a-new-cmu-study-confirms/The LLMs are just somewhere between an averaging and a lossy compression of everything on GitHub. There's nothing about the current paradigm of "AI" that is going to somehow do better than just rehashing that training set but with the inclusion of various classes of errors.
I think it's better to view it as spicy search rather than any form of intelligence.
Yeah that's mostly what i use it for. A way to search for things that i can't describe well enough for a traditional search engine so I can find out at a glance what it's name is and if it's valid for my situation. If it is then I go look up documentation. Any time I've stayed in the LLM past that I eventually go down a rabbit hole of wrong ideas that aren't always obvious until you get a bit too deep and you've wasted an hour with an incorrect solution.
The only thing I have found it useful for is book recommendations. I like this book and that book, what other types of books are like these?
Yeah, I do believe it's a good tool for search, just with the caveat that if it can't find an answer it makes one up or otherwise kinda just fills in little missing details with noise.
If you treat it as spicy search it works pretty great though.
As I'm slowly evolving my own flavour of spec driven development, I'm starting to think about the generated code as a secondary artefact where main quality criteria is that it's doing what it needs to and it's covered with tests.
I guess my current analogy is that I don't care about how readable or dry is the assembly code generated by compiler.
I have the specifications and the working code with tests. I can always regenerate it if I need to.
But. I still read the produced code, steer the design and correct the obvious blunders. No vibes.
Might want to read on TDD, it's been around since last the last millennium (OK 1999 according to Wikipedia, point is, it's not new).
I know a lot of people hate but this AI stuff still isn’t great but it will get better. Each generation of programming languages adds syntax and convenience. AI code will likely get to the point where it is just a higher level language. The only benefit I’m seeing is that if used very carefully I can make more complex projects with fewer team members. And where there was zero documentation there’s at least SOME documentation.
While noobs and managers are excited that the input language to this compiler is English, English is a poor language choice for many reasons.
Yep, it is a poor choice today.
Like all things, it will likely improve. I see a world where a pseudo-code format and some standard start to form.
Until then, it is the wild west, and I fear some people may die from the misuse of these vibe coding tools. But they aren’t necessarily useless.