Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

futurism.com

Post reply on HN
Print view View original

61–70 of 174 posts

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#61

Post by zitterbewegung »

jasonthorsness wrote: post #19

Half of the work is specification and iteration. I think there’s a focus on full SWE replacement because it’s sensational, but we’ll more end up with SWE able to focus on the less patterned or ambiguous work and made way more productive with the LLM handling subtasks more efficiently. I don’t see how full SWE replacement can happen unless non-SWE people using LLMs become technical enough to get what they need out of…

If the llm can’t find me a solution in 3 to 5 tries while I improve the prompt I fall back to mire traditional methods and or use another model like Gemini.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#62

Post by aprilthird2021 »

rurp wrote: post #36

I recently had to do a one-off task using SQL in a way that I wasn't too familiar with. Since I could explain conceptually what I needed but didn't know all the right syntax this seemed like a perfect use case to loop in Claude. The first couple back and forths went ok but it quickly gave me some SQL that was invalid. I sent back the exact error and line number and it responded by changing all of the aliases but repe…

I do something like this every day at work lol. It's a good base to start with, but often you'll eventually have to Google or look at the docs to see what it's messing up

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#63

Post by sanxiyn »

jasonthorsness wrote: post #19

Half of the work is specification and iteration. I think there’s a focus on full SWE replacement because it’s sensational, but we’ll more end up with SWE able to focus on the less patterned or ambiguous work and made way more productive with the LLM handling subtasks more efficiently. I don’t see how full SWE replacement can happen unless non-SWE people using LLMs become technical enough to get what they need out of…

Everyone is a typist now, so I don't think it is farfetched that everyone is a SWE in the future.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#64

Post by DarkmSparks »

anandnair wrote: post #38

Coding, especially the type mentioned in the article (building an app based on a specification)—is a highly complex task. It cannot be completed with a single prompt and an immediate, flawless result. This is why even most software projects (built by humans) go through multiple iterations before they work perfectly. We should consider a few things before asking, "Can AI code like humans?": - How did AI learn to code?…

>How did AI learn to code?

It didn't, it's just very good at copying already existing code and tweeking it a bit.

>Did AI receive mentoring from an experienced senior

It doesnt even comprehend what an experienced senior is, all it cares about is how frequently certain patterns occurred in certain circumstances.

>Did the AI learn through hands-on coding or just by reading Stack Overflow?

it "learnt" by collecting a large database of existing code, most of which is very low quality open source proofs of concept, then spits out the bits that are probably related to a question.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#65

Post by brookst »

ipython wrote: post #50
pzo wrote: post #35

> The models weren't allowed to access the internet How many software developers could solve most even simple programming problems (except 'Hello world') with zero shot style (you write in notepad then can compile only once and execute once) without access to internet (stackoverflow, google search, documentation), tools (terminal, debugger, linter, cli)? I think then it's not the best comparison to make any judgement…

What would searching the Internet provide the models that they don’t already have? Most likely data sources such as stack overflow, documentation on the language it’s targeting, and a variety of relevant forum posts are already part of its training set. Unless someone else came along and said “here’s how to solve x problem step by step”, I don’t see how additional information past its cutoff point would help. (Perhap…

Why do programmers search for specific questions rather than always relying on their inherent knowledge?

I’m a crappy hobbyist programmer but for me it is useful to see if someone has implemented exactly what I need, or debugged the problem I’m having. I don’t think it’s reasonable to expect programmers or LLMs to know everything about every library’s use in every context just from first principles.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#66

Post by simonw »

I find the framing of this story quite frustrating.

The purpose of new benchmarks is to gather tasks that today's LLMs can't solve comprehensively.

It an AI lab built a benchmark that their models scored 100% on they would have been wasting everyone's time!

Writing a story that effectively says "ha ha ha, look at OpenAI's models failing to beat the new benchemark they created!" is a complete misunderstanding of the research.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#67

Post by hackit2 »

rurp wrote: post #36

I recently had to do a one-off task using SQL in a way that I wasn't too familiar with. Since I could explain conceptually what I needed but didn't know all the right syntax this seemed like a perfect use case to loop in Claude. The first couple back and forths went ok but it quickly gave me some SQL that was invalid. I sent back the exact error and line number and it responded by changing all of the aliases but repe…

Makes perfect sense why it couldn't answer your question, you didn't have the vocabulary of relational algebra to correctly prime the model. Any rudimentary field have their own corpus vocabulary to express ideas and concepts specific to that domain.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#68

Post by jameslk »

rurp wrote: post #36

I recently had to do a one-off task using SQL in a way that I wasn't too familiar with. Since I could explain conceptually what I needed but didn't know all the right syntax this seemed like a perfect use case to loop in Claude. The first couple back and forths went ok but it quickly gave me some SQL that was invalid. I sent back the exact error and line number and it responded by changing all of the aliases but repe…

I had to do something similar with BigQuery and some open source datasets recently.

I had bad results with Claude as you mentioned. It kept hallucinating parts of the docs for the open datasets, coming up with nonsense columns. Not fixing errors when presented the error text and more context. I had a similar outcome with 4o.

But I tried the same with o1 and it was much better consistently, with full generations of queries and alterations. I fed it in some parts of docs anytime it struggled and it figured it out.

Ultimately I was able to achieve what I was trying to do with o1. I’m guessing the reasoning helped, especially when I confronted it about hallucinations and provided bits of the docs.

Maybe the model and the lack of CoT could be part of the challenge you ran into?

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#69

Post by ripped_britches »

tkgally wrote: post #56
jasonthorsness wrote: post #19

Half of the work is specification and iteration. I think there’s a focus on full SWE replacement because it’s sensational, but we’ll more end up with SWE able to focus on the less patterned or ambiguous work and made way more productive with the LLM handling subtasks more efficiently. I don’t see how full SWE replacement can happen unless non-SWE people using LLMs become technical enough to get what they need out of…

> unless non-SWE people using LLMs become technical enough to get what they need out of them Non-SWE person here. In the past year I've been able to use LLMs to do several tasks for which I previously would have paid a freelancer on Fiverr. The most complex one, done last spring, involved writing a Python program that I ran on Google Colab to grab the OCR transcriptions of dozens of 19th-century books off the Interne…

This is a good anecdote but most software engineering is not scripting. It’s getting waist (or neck) deep in a large codebase and many intricacies.

That being said I’m very bullish on AI being able to handle more and more of this very soon. Cursor definitely does a great job giving us a taste of cross codebase understanding.

Re: OpenAI Researchers Find That AI Is Unable to Solve Most Coding Problems

#70

Post by tsimionescu »

anandnair wrote: post #38

Coding, especially the type mentioned in the article (building an app based on a specification)—is a highly complex task. It cannot be completed with a single prompt and an immediate, flawless result. This is why even most software projects (built by humans) go through multiple iterations before they work perfectly. We should consider a few things before asking, "Can AI code like humans?": - How did AI learn to code?…

LLMs can fundamentally only do something similar to learning in the training phase. So by the time you interact with it, it has learned all it can. The question we then care about is whether it has learned enough to be useful for problem X. There's no meaningful concept of "how intelligent" the system is beyond what it has learned, no abstract IQ test decoupled from base knowledge you could even conceive of.
Post reply on HN