Hacker Newsnew | past | comments | ask | show | jobs | submit | lqstuart's commentslogin

Not long ago I started having an issue with my eye. I called around and they said I should get seen ASAP, same day if possible, but it wasn’t worth the ER and it was a five day wait for an appointment.

I was pretty freaked out. During that time, I tried diagnosing it with AI. When I finally got to the appointment, the actual doctor sat down, looked at all the unremarkable images, asked me one (1) question, ordered another image and diagnosed the issue. When I looked back, in all that time, the AI had mentioned it exactly one time early on, ruled it out immediately based on a flawed understanding of the symptoms, and never brought it up again.

Just my anecdotal evidence, but I’d never trust any AI on its own. My doctor can use it if they want, I can’t.


It’s the ER. People aren’t always in a position to “chat” when they go there.

You think current ER people work in complete silence? No words uttered?

You think that they have “days leading up to consultation”? Please don’t be so disingenuous; I’m sure you know exactly what the person you’re replying to meant.

> I’m sure you know exactly what the person you’re replying to meant.

No.

There are a lot of different modus operandi, and you can always find an outlier.

> Please don’t be so disingenuous;

Ditto


Too bad they don't care so much about factory worker safety or slave labor

I don't think the bake-off to decide which superpower has the worst human rights record is going to land where you want it to. Hint, they both suck.

FWIW, I'll take the one not dropping bombs to keep their BFF happy, boosting right-wing shitheads, threatening to invade their real allies and slapping dumb tariffs on everyone.


They both suck, but one of them literally harvests organs from political prisoners.

I’m honestly torn on which one I’d pick, but there’s a TON of likely state-sponsored pro-China propaganda on the internet, so I consider it a patriotic duty to push back for the sole reason that we can still freely talk shit about the one (for the time being, as long as you don’t mention the blessed martyr Charlie Kirk), whereas the other blocks the internet and imprisons people for dissent.


The country I live in has a specific cutout for allowing slavery for incarcerated people.

I had a TI-89 in high school. I was that kid.

The “$500B AI labs” are also predominantly ~200 person Chinese teams…

Buried pretty deep in the article

> “The raw output of ChatGPT’s proof was actually quite poor. So it required an expert to kind of sift through and actually understand what it was trying to say,” Lichtman says. But now he and Tao have shortened the proof so that it better distills the LLM’s key insight.

I guess “ChatGPT came up with a novel approach to a problem that later turned out not to be totally stupid and terrible for once” isn’t as catchy of a headline


I understood this to mean that the ChatGPT output was technically correct, just hard to understand.

I haven't reviewed it myself, but when a mathematician calls a proof "quite poor" and experts have to "sift through" it, I would understand that to mean that it's technically incorrect. Errors like "This statement isn't correct, but it points towards a weaker statement that is, and the subsequent steps can be rebuilt on top of the weaker statement" are pretty common in output from both LLMs and math students.

No - It likely means that the proof was meandering, and had lots of additional pointless steps.

Good/bad is orthogonal to correct/incorrect.

I wouldn't expect a hand-crafted proof by an amateur to be much different.

Depends. I reckon a proof by an amateur would either be worthless because it demonstrates no understanding whatsoever or significantly better because they actually understand the proof.

LLM produced texts are often in a weird area where the quality of the content and the quality of the writing have very little to do with one another.


I don't think it's true that all amateurs have no understanding whatsoever. Amateurs have proven things before, and they've also wasted mathematicians time with wrong proofs.

I'm not saying none of them have any understanding, I'm saying the ones who understand would write better proofs.

Wouldn’t the expectation for ChatGPT be that it presents a well refined report, rather than hand crafted proof / notes?

Because from what I gather, they basically had to go through the equivalent of a pile of notes to find the crux.


Yeah, I also expect better than amateur-level redaction from a system that OpenAI marketing sells as a team of PhD in your pocket.

How many hand-crafted amateur proofs do you read in a month? If the answer is close to zero then what are your expectations actually driven by?

My understanding of the world, human nature and human learning?

Doesn’t this mean the expert solved the problem while trying to devine the tea leaves provided by chatgpt?

How often are humans initial key insights, also sublimely distilled and beautiful.

This is like comparing someone's first draft, with a final published paper.


That should be buried, I agree 100% with their headline and structure over yours.

For comparison, if the amateur did it by hand but the result was sloppy to read, would you prefer "Amateur solves an Erdos problem" or "Amateur came up with a novel approach to a problem that later turned out not to be totally stupid and terrible for once"?


DeepSeek also seems capable of solving it. In under 20 minutes

https://chat.deepseek.com/share/nyuz0vvy2unfbb97fv

I guess we should test across other LLMs too


Do you have any idea if this is correct?

There should be zero expectation that the solution is "novel." It could not have produced any of it were it not in it's training data set.

This is simply evidence that our search tools and academic publishing are completely broken and not at all evidence that a machine "thought up a novel solution."

Humans constantly anthropomorphize their environment. To their detriment.


If that’s the case, are you saying the proof was part of the training set?

A lot of novelty is just gluing approaches together and reporting what sticks.


Yes.

A lot of brute force methods are the most inefficient means of solving a problem.

Language Models are a sign that our current human information infrastructure and access methods are completely wrong.


Arguably, since everything resolves to an axiom, isn’t the solution in any training set ?

The order and combination is what makes it special.

Is current human information access methods wrong, or do we just synthesize data in a way that is inefficient for this sort of problem solving ?


> since everything resolves to an axiom

This isn't true. There are solutions that are beyond apparent reason and logic. This is what a "breakthrough" is.

> The order and combination is what makes it special.

Given an infinite amount of time a team of monkeys will produce Shakespeare. Is that "special?" Perhaps we should leave some room for _how_ those combinations happen and how efficient they are.

> Is current human information access methods wrong

They are wrong. The largest search company is also the largest advertiser. I'm surprised that anyone either fails to apprehend this or pretends not to.


What do you call the law that you violate when you vibe code an entire website for "List of 'laws' of software engineering" instead of just creating a Wikipedia page for it


"Creating a Wikipedia page" is a weird suggestion. In 2026, it's actually not possible to create a Wikipedia page unless you're already a deep expert in Wikipedia culture.

(Wikipedia nerds often say "No, anyone can create a page as long as they follow the 137 guidelines!" This is a prank- Wikipedia admins will delete your article no matter how many guidelines it follows)


Even some 15y ago it was impossible to add web links to communities, even though other web links to similar communities were already in the web links section, because some people weaponized wiki as a moat.


Slop’s Law

If it can be slopped, it will be slopped.


Sturgeon's Law (2026): 99% of everything is crap or slop


My company has four (4) vibe-coded dashboards to monitor AI tool usage.

We have made no revenue, let alone profit from any AI feature. However, some curiously under qualified people have been hired into new “AI” themed roles with seven or eight figure comp, and we seem to be preparing for major layoffs in the next 30-60 days. Presumably those new roles will be safe.


I switched from Windows to Linux because I got a Steam Deck, which caused me to realize that the only games in my library that don't also run flawlessly on Linux are the ones that have invasive anticheat that I'm really not comfortable installing.

Having to enable TPM or device integrity or whatever it is on my own computer just to run my own games is just too much power to hand to some garbage corporation that shits on its users. Rubbed me so far the wrong that way that I gave it up. The fact that Win 11 is no longer just an easy and hands-off solution that "just works" but is bloated with dark patterns and "AI" bullshit certainly helped cement the decision.


Have a Linux gaming machine with bazzite. Fantastic! So far any game I care about works and with AMD drivers it’s even stable..


you mean the $5 billion they've generated off of the $73 billion they've raised?


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: