HN user
fortis fortuna adiuvat
I have always been saying: don't be optimistic about AI. Software people prepared their end with this optimism without knowing how the economy works. "AI will make our jobs easier, I will have more time with my family," blah blah. Now we will see more news like that in the coming years. I am okay with supporting AI for curing cancer, but I am against supporting AI for disrupting the white-collar economy without a proper government plan.
It's showcase more than being take home assignment. I couldnt understand what the task is ,only performance comparisons between their LLM
https://news.ycombinator.com/item?id=46550836 Another view on that.
It's hard to cut through the AI hype when there are billions of dollars at stake. I usually trust negative comments more, as long as the person isn't trying to sell a course. Even though Terence Tao is a respected scientist, I wonder if his recent comments are driven by a need for funding due to federal cuts. I’ve had similar experiences with LLMs—whenever I ask them about hard math or RL theory, they almost always give me the wrong answers.
I am not mathematician, so it's hard to understand the comment. Thank you for your warning. I will be more careful.
I read how GPT-5 contributed to proof. It is not fully solved by GPT-5 instead assisted. For more look here https://www.math.columbia.edu/~msawhney/Problem_848.pdf
This is response from mathematician: "This is quite something, congratulations to Boris and Aristotle!
On one hand, as the nice sketch provided below by tsaf confirms, the final proof is quite simple and elementary - indeed, if one was given this problem in a maths competition (so therefore expected a short simple solution existed) I'd guess that something like the below would be produced. On the other hand, if something like this worked, then surely the combined talents of Burr, Erdős, Graham, and Li would have spotted it.
Normally, this would make me suspicious of this short proof, in that there is overlooked subtlety. But (a) I can't see any and (b) the proof has been formalised in Lean, so clearly it just works!
Perhaps this shows what the real issue in the [BEGL96] conjecture is - namely the removal of 1 and the addition of the necessary gcd condition. (And perhaps at least some subset of the authors were aware of this argument for the easier version allowing 1, but this was overlooked later by Erdős in [Er97] and [Er97e], although if they were aware then one would hope they'd have included this in the paper as a remark.)
At the moment I'm minded to keep this as open, and add the gcd condition in the main statement, and note in the remarks that the easier (?) version allowing 1 and omitting the gcd condition, which was also asked independently by Erdős, has been solved."
The commentator is saying: "I can't believe this famous problem was solved so easily. I would have thought it was a fake proof, but the computer verified it. It turns out the solution works because it addresses a slightly different set of constraints (regarding the number 1) than what Erdős originally struggled with. (Generated by Gemini)
I started fully coding with Claude Code. It's not just vibe coding, but rather AI-assisted coding. I've noticed there's a considerable decrease in my understanding of the whole codebase, even though I'm the only one who has been coding this codebase for 2 years. I'm struggling to answer my colleagues' questions.
I am not defending we should drop AI, but we should really measure its effects and take actions accordingly. It's more than just getting more productivity.
LLMs can do small things well, but you must use small parts to form big picture.
it is not that easy, there is lazy loading in the page that is triggered by scroll of specific sections. You need to find clever way, no way to scrape with bs4, so tough with even selenium.
Yesterday, I was asked to scrape data from a website. My friend used ChatGPT to scrape data but didn't succeded even spent 3h+. I looked website code and understand with my web knowledge and do some research with LLM. Then I described how to scrape data to LLM it took 30 minutes overall. The LLM cant create best way but you can create with using LLM. Everything is same, at the end of the day you need someone who can really think.
I mean test-time scaling coming to end, there are many open rooms for next thing.
test time scaling is based on methods from pre-2020. If you look details of modern LLMs its pretty small prob to encounter method from 2020+(ROPE,GRPO). I am not saying IMO is not impressive, but it is not breakthrough, if they said they used different paradigm then test-time scaling I would say breakthrough.
We are not close to solving IMO with publicly known methods. The point here is not method rather computation power. You can solve any verifiable task with high computation, absolutely there must be tweaks in methods but I don't think it is something very big and different. Just OAI asserted they solved with breakthrough.
Wait for self-adapting LLMs. We will see at most in 2 years, now all big tech are focusing on that I think.
I don't remember as a big fan of DeepSeek.
IMO is not breakthrough, if you craft proper prompts you can excel imo with 2.5 Pro. Paper : https://arxiv.org/abs/2507.15855. Google just put whole computational power with very high quality data. It was test-time scaling. Why it didn't solve problem 6 as well?
Yes, it was breakthrough but saturated quickly. Wait for next breakthrough. If they can build adapting weights in llm we can talk different things but test time scaling coming to end with increasing hallucination rate. No sign for AGI.
Seems LLMs really hit the wall.
That's what I meant. I don't think boss wants you to pay same money with less work time.
If an AI can replace these repeated tasks, I could spend more time with my fiancé, family, friends, and dog, which is awesome, and I am looking forward to that.
I could not understand this optimism, aren't we living in a capitalist world ?
This is bad use of AI, we spend our compute to make science faster. I am pretty confident computational cost of this will be maybe 100x of chatgpt query. I don't want to think even environmental effects.
We didn't even reach AGI and no sign with LLMs.
If AI soon becomes good enough at building software on its own, software engineering as we know it is dead. I have no interest in becoming a glorified project manager, orchestrating AI agents all day long. If it does happen, I am now competing with anyone who can type a prompt. I’m not betting my career on being slightly better at prompting than millions of others.
The view I most agree with this discourse. That's why I am not enthusiastic about AI
Appreciated
I will not be commenting on any self-reported AI competition performance results for which the methodology was not disclosed in advance of the competition.
Google also joined IMO, and got gold prize.
https://x.com/natolambert/status/1946569475396120653
OAI announced early, probably we will hear announcement from Google soon.
I think from Canada team someone solved it but among all, its very few
I mean progress speed, few months ago they released o3 it has 16 pt in imo 2025
Progress is astounding. Recently report published about evaluation of LLMs on IMO 2025. o3 high didn't even get bronze.
Waiting for Terry Tao's thoughts, but these kind of things are good use of AI. We need to make science progress faster rather than disrupting our economy without being ready.
I was really shocked when I first see this but yes it's in training data. Not thinking feature.
How will you figure out or solve hallucinated assembly code ?
omg, my ai agent did nil dereferencing, it seems it's trying to implement backdoor to my system so that it will crash my server.
If you invent someting better, you can get reward, lots of examples out there.