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

1 you are massively assuming less than linear improvement, even linear over 5 years puts LLM in different category

2 more efficient means need less people means redundancy means cycle of low demand


1 it has nothing to do with 'improvement'. You can improve it to be a little less susceptible to injection attacks but that's not the same as solving it. If only 0.1% of the time it wires all your money to a scammer, are you going to be satisfied with that level of "improvement"?

> You can improve it to be a little less susceptible to injection attacks

That’s exactly the point the rapid rate of improvement is far form slow polish in 10 years it will be everywhere doing everything


I think you missed the other half of the sentence. It's not converging on 'immune' no matter how fast it improves.

OK. Let's take what you've stated as a truth.

So where is the labor force replacement option on Anthropic's website? Dario isn't shy about these enormous claims of replacing humans. He's made the claim yet shows zero proof. But if Anthropic could replace anyone reliably, today why would they let you or I take that revenue? I mean they are the experts, right? The reality is these "improvements" metrics are built in sand. They mean nothing and are marketing. Show me any model replacing a receptionist today. Trivial, they say, yet they can't do it reliably. AND... It costs more at these subsidized prices.


Why is the bar replacing a receptionist ? At the low end It will take over tasks and companies will need less people, at the top end it will take over roles. What’s the point you are making, if it can’t do bla now it never will ?

Then define the bar. You're OK with all of these billionaires just saying "we're replacing people in 6-60 months" with no basis, no proof, no validation? So the onus is now on the people who challenge the statement?

Why is the bar not even lower you ask? Well I guess we could start with replacing lying, narcissistic CEOs.


LLMs haven't been improving for years.

Despite all the productizing and the benchmark gaming, fundamentally all we got is some low-hanging performance improvements (MoE and such).


Wow you are making a point of everything will be ok using farming ! Farming is struggling consolidated to big big players and subsidies keep it going

You get layed off and spend 2-3 years migrating to another job type what do you think g that will do to your life or family. Those starting will have a paused life those 10 fro retirement are stuffed.


True but who is porting the top 20 languages to risc-v with the same performance

We don’t lack the technology to limit scrapers, sure it’s an arms race with AI companies with more money than most. Why can’t this be a legal block through TOS

How is the publisher supposed to fund their operations let along make a profit. How about a 1 year lock on the archive pages. There are many ways of keeping that record but not taking views undermining the business model

The same way they did back in the day, where libraries still existed that allowed people to read newspapers for free.

I kind of doubt that internet archive is really taking very much business away from them. Its a terrible UI to read the daily news.


The LWN model feels practical here:

> We ask that you grant LWN exclusive rights to publish your work during the LWN subscription period - currently up to two weeks after publication.

News is valuable when it is timely, and subscribers pay for immediate access.

https://lwn.net/op/AuthorGuide.lwn


> How is the publisher supposed to fund their operations let along make a profit.

There used to be plenty newspapers sponsored by wealthy industrialists; the latter would cover the former's gaps between the costs and the sales, the former would regularly push the latter's political agenda.

The "objective journalism" is really quite a late invention IIRC, about the times of WW2.


Objectivity was already a principle in the 1890s.

https://en.wikipedia.org/wiki/Journalistic_objectivity

"To give the news impartially, without fear or favor." — Adolph Ochs, 1858-1935

Objectivity is the default state of honest storytelling. If I ask what happened ? and somebody only tells the parts that suit an agenda, they have not informed me. The partisan press exists, because someone has a motive to deviate from the natural expectation of fair story telling and story recounting.


> Objectivity is the default state of honest storytelling. If I ask what happened ? and somebody only tells the parts that suit an agenda, they have not informed me.

Already at the level of what stories are covered you have made choices about what's important or not.

Your newspaper not covering your neighbors lawsuit against the city against some issue because they find it to be "not important" is already a viewpoint-based choice

A newspaper presenting both sides on an issue (already simplifying on the "there are two sides to an issue" thing) is one thing. Do you also have to present expert commentary that says that one side is actually just entirely in bad faith? Do you write a story and then conclude "actually this doesn't matter" when that is the case?

There are plenty of descriptions that some people would describe as fair story telling and others would describe as a hit piece. Probably for any article on any controversial topic written in good faith you are likely able to find some people who would claim it's not.

I think it's important to acknowledge that even good faith journalism is filled with subjectivity. That doesn't mean one gives up, you just have to take into account the position of the people presenting information and roll with that.


You make it sound like bias is completely relative and undecidable. But there is a clear line journalists can cross - if they're intentionally misleading their reader, that's bias. It's qualitatively different from neglecting to cover a story or not finding a suitable expert or whatever. It's intentional deception because they want the readers to have wrong knowledge. And they do it all the time.

It's a great question, but they didn't seem to have a problem with this before AI, so I have to assume that the presence of a free available copy wasn't really impacting their revenue.

If an independent press is critical to open societies, perhaps some sort of citizen directed funding is needed to maintain independence from both capital and government?

Maybe it would be better if these news operations had to find better ways to sustain themselves than the current paradigms. Also, the internet archive is not the only archive, and there will be more. This ins't something they can really stop.

Reconfigure human society so that services like news don't need to make a profit and still remain credible.

I must be a Luddite, how do you have a model working for 12 hours on a problem. Mine is ready with an answer and always interrupts to ask confirmation or show answer

That's on the harness - the device actually sending the prompt to the model. You can write a different harness that feeds the problem back in for however long you want. Ask Claude Code or Codex to build it for you in as minimal a fashion as possible and you'll see that a naïve version is not particularly more complex than `while true; do prompt $file >> file; done` (though it's not that precisely, obviously).

How will these investors get their money. The AI companies either have a competitor invent AGI which means all other AI companies are worth nothing or they themselves invent AGI in which case a whole bunch of companies go broke including the investors themselves.

This money isn’t never going to be returned


A few of these, likely openai and anthropic, will IPO, and the early stage investors will get their cash back then. Others will crash, maybe also openai and anthropic. Other investors in smaller firms may get cash back when the companies are swallowed up by Google, OpenAI, Anthropic, Apple, etc., when the Great and Inevitable Consolidation occurs. I think short term there's a ton of money to be made by investing in these companies. In the long run I don't see how they survive, since nobody in the big leagues has a real technological advantage over anybody else so stiff competition will keep prices low. Even if they cartelize, a startup could beat them on price since there's not much preventing anyone with enough capital from building their own "frontier" models now, except GPU shortages... Talent of course, but there's plenty of smart people out there who would take a big paycheck to do it.

The output is a feature, a bug fix, a product.

Correct me if I’m wrong aren’t they the innovators of multiple things like skills sub agents mcp and whatever this memory thing is agents files

Seriously they are the apple iPhone or AWS of LLM a decade or so ago.


Why doesn’t you ask it and find out ;)


Because the model doesn't know but will happily tell a convincing lie about how it works.


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

Search: