[Klaster_1]

This article matches my experience as well. Chatting with LLM has helped me to crystalize ideas I had before and explore relevant topics to widen the understanding. Previously, I wouldn't even know where to begin with when getting curious about something, but ChatGPT can tell you if your ideas have names, if they were explored previously, what primary sources there are. It's like a rabbit hole of exploring the world, a more interconnected one where barriers of entry to knowledge are much lower. It even made me view things I previously thought of as ultra boring in different, more approachable manner - for example, I never liked writing, school essays were a torture, and now I may even consider doing that out of my own will.

[snek_case]

In the early 2000s Wikipedia used to fill that role. Now it's like you have an encyclopedia that you can talk to.

What I'm slightly worried about is that eventually they are going to want to monetize LLMs more and more, and it's not going to be good, because they have the ability to steer the conversation towards trying to get you to buy stuff.

[pmarreck]

> they are going to want to monetize LLMs more and more

Not only can you run reasonably intelligent models on recent relatively powerful PC's "for free", but advances are undoubtedly coming that will increase the efficient use of memory and CPU in these things- this is all still early-days

Also, some of those models are "uncensored"

[vjk800]

Can you? I imagine e.g. Google is using material not available to the public to train their models (unsencored Google books, etc.). Also, the chat bots, like Gemini, are not just pure LLMs anymore, but they also utilize other tools as part of their computation. I've asked Gemini computationally heavy questions and it successfully invokes Python scripts to answer them. I imagine it can also use other tools than Python, some of which might not even be publicly known.

I'm not sure what the situation is currently, but I can easily see private data and private resources leading to much better AI tools, which can not be matched by open source solutions.

[croon]

While they will always have premiere models that only run on data center hardware at first, the good news about the tooling is that tool calls are computationally very minimal and no problem to sandbox/run locally, at least in theory, we would still need to do the plumbing for it.

So I agree that open source solutions will likely lag behind, but that's fine. Gemini 2.5 wasn't unusable when Gemini 3 didn't exist, etc.

[fennecbutt]

Yes, because local models can run Internet search tools. Even the big boys like openai etc I prefer the results quality when it's made a search - and they seem to have realised this too, the majority of my queries now kick off searches.

[OGEnthusiast]

How do you verify the models you download also aren't trying to get you to buy stuff?

[tvink]

I guess you.. ask them a bunch of recommendations? I would imagine this would not be incredibly hard to test as a community

[ben_w]

Before November 30, 2022 that would have worked, but I think it stopped being reliable sometime between the original ChatGPT and today.

As per dead internet theory, how confident are we that the community which tells us which LLM is safe or unsafe is itself made of real people, and not mostly astroturfing by the owners of LLMs which are biased to promote things for money?

Even DIY testing isn't necessarily enough, deceptive alignment has been shown to be possible as a proof-of-concept for research purposes, and one example of this is date-based: show "good" behaviour before some date, perform some other behaviour after that date.

[awesome_dude]

Proudly bought to you by Slurm

[ReedorReed]

Which free models do you recommend?

[Klaster_1]

One of the approaches to this that I haven't seen being talked about on HN at all is LLM as public infrastructure by the government. I think EU can pull this off. This also addresses overall alignment and compute-poverty issue. I wouldn't mind if my taxes paid for that instead of a ChatGPT subscription.

[Nevermark]

This is not a good idea at all.

Government should not be in a position to directly and pervasively shape people’s understanding of the world.

That would be the infinite regress opposite of a free (in a completely different sense) press.

A non-profit providing an open data and training regime for an open WikiBrain would be nice. With standard pricing for scaled up use.

[squidbeak]

> Government should not be in a position to directly and pervasively shape people’s understanding of the world.

You disagree with national curricula, state broadcasters, publicly funded research and public information campaigns?

[cloverich]

Many Americans these days absolutely do disagree with all of those things. Educated ones. Theres simply a short circuit belief based pathway in peoples brains that bypasses everything rational on arbitrary topics.

Most of us used to see it as isolated to religion or niche political points, but increasingly everything os being swept into the "its political" camp.

[Nevermark]

Given “national curricula” of a dominant democratic country are undergoing a politically motivated change, starting with significant web materials, and moving into education …

Do you prefer the previous narratives? The latter? Or whatever you are told?

And that is the risk of relatively static information.

What if your information source was interactive, adaptive and motivated? And could record and report on your interactions and viewpoints?

[gloomyday]

I've heard once that Americans distrust their government and trust their corporations, while Europeans distrust their corporations and trust their government. I honestly think that governments already has a huge role in shaping people's understanding of the world, and that's GREAT on good democratic countries.

What I find really weird is that I am stopping believing in the whole idea of free press, considering how the mainstream media is being bought by oligarchs around the globe. I think this is a good example of the erosion of trust in institutions in general. This won't end well.

Your idea of letting it be run by a non-profit makes me believe that you also don't trust institutions anymore.

[Nevermark]

I can’t say I have no trust for any institutions.

But my trust depends on each institutions choices. Just as my trust in people varies based on their records.

Mostly, I trust everyone to be who they show themselves to be. Which can lean one way or the other, or be very mixed across different dimensions.

But, yes, governments and corporations which are steadily centralizing power are inherently untrustworthy. As they at best, are making us all vulnerable and less powerful as individuals. Meaning they are decreasing our freedom, not increasing it.

[navane]

Instead, we should let capitalism consolidate all power in the hands of the few, and let them directly and pervasively shape people's understanding of the world.

How would a non profit even be funded? That would just be government with extra steps.

No, capitalism giveth the LLMs and capitlism taketh the sales.

[Nevermark]

Were you responding to someone else?

For answers, just re-read my comment. Or, this:

1) Avoiding centralization is exactly why government shouldn’t do this.

2) Why did you raise a false dichotomy of government vs. commercial centralization?

I proposed an open solution, which is non-commercial with decentralized control.

3) Funding?: Have you heard of Wikipedia?

People often donate to prominent tools they use.

And, as I pointed out, there is an even more reliable source.

The necessity for scaled automated access creates an inevitable need for uniform openly set pricing for scaled up access.

I nice case where non-profit open access is naturally self-funded.

[xvokcarts]

That's because a government-run LLM would be like government-run media.

High inflation? No, the government LLM will tell us we're doing great.

[attila-lendvai]

this assumes that "the government" is "us" and not "them"...

[michaelbrave]

I sort of already had an experience where it did kinda. I was consulting with it potential fashion choices to upgrade my work uniform, to look professional but still creative, basically to look more like a creative director. It recommended brands, colors, styles etc. Then I was asking about eyeglass frames showed it three pictures, described my facial features and it was like "you have to buy this one now" more enthusiastic than expected. It wasn't ads or anything but there was a bit of salesyness in there.

[attila-lendvai]

or more generally than just ads: make you believe stuff that makes you act in ways that is detrimental to you, but benefitial to them (whoever sits in the center and can control and shape the LLM).

i.e. the Nudge Unit on steroids...

care must be taken to avoid that.

[anon_anon12]

I can envision this as being in the boots of Truman from Truman Show where some advertisement is thrown at your face randomly

[idiotsecant]

It's also inevitable that better and better open source models will be distilled as frontier models advance.

[snek_case]

I agree. I think the local models you can run on the "average computer" are not quite good enough yet, but I have hope that we will see much better small local models in the future.

[resonious]

Right, this is what happened with search engines. And "SEO for LLMs" is already a thing.

[deadbabe]

Enshittification is always inevitable in a capitalist world, but not always easy to predict how it will happen.

[afro88]

I'm not great with math beyond high school level. But I am very interested in, among many things, analog synthesiser emulations. The "zero delay filter" was a big innovation in the mid 2000s that led to a big jump in emulation accuracy.

I tried to understand how they work and hit a brick wall. Recently I had a chat with an LLM and it clicked. I understand how the approximation algorithm works that enables solving for the next sample without the feedback paradox of needing to know it's value to complete the calculation.

Just one example of many.

It's similar to sitting down with a human and being able to ask questions that they patiently answer so you can understand the information in the context of what you already know.

This is huge for students if educational institutions can get past the cheating edge of the double edged sword.

[attila-lendvai]

educational institutions became pretty much obsolete with the advent of the internet (i.e. marginalized the cost of the flow of information).

what we need (if anything besides reputation tracking), is (maybe) a separate institution for testing and issuing diplomas... which, BTW, can be trusted more with QA than the very producers themselves.

producers = QA has always been such a contradiction of schools...

[fatherwavelet]

audio DSP is a great example. Especially because the proof is in the output if you are then able to make a cool sounding analog filter emulation that you would not have otherwise.

I disagree on the students in educational institutions though. This is the biggest thing for the autodidact who doesn't have the privilege of the educational institution. I think of time, money and effort that would be needed to talk to a professor one on one about analog filter emulation. It is not happening for me.

There are also societal/social high pass filters to even get in the door for these educational institutions. I would just get filtered out anyway. It seems to me in time that entire concept will simply become null and void.

[whatever1]

LLM as a rubber duck is a great use case.

[NitpickLawyer]

Especially with the great unlock that was made possible by large (usable) context windows. You can now literally "throw the book" at an LLM and get grounded stuff that is at or above what you could do yourself. But it can be done on-demand, on almost every subject there is, at lower and lower cost.

[montag]

Wasn’t it George Orwell who said “writing is thinking”?

[Klaster_1]

That sounds true when you have already internalized the idea. But my environment never even suggested in can be approached this way. My school didn't teach me how to write, we just had to. Uni didn't explain this either and that was part of the reason why I dropped out. You can't make progress when you don't know what questions to ask and nobody sees your struggles and provides guidance.

[gradus_ad]

I think the best existing "product" analogy for LLM's is coffee.

Coffee is a universally available, productivity enhancing commodity. There are some varieties certainly, but at the end of the day, a bean is a bean. It will get the job done. Many love it, many need it, but it doesn't really cost all that much. Where people get fancy is in all the fancy but unnecessary accoutrements for the brewing of coffee. Some choose to spend a lot on appliances that let you brew at home rather than relying on some external provider. But the quality is really no different.

Apparently global coffee revenue comes out to around $500B. I would not be surprised if that is around what global AI revenue ends up being in a few years.

[teiferer]

> Coffee is a universally available, productivity enhancing commodity

The analogy carries further than you intended. If you have never reached addiction stage, then there is no factual productivity enhancement. "But I'm so much less productive if I haven't had my morning coffee" Yeah, because you have an addiction. It sounds worse than it is, if you just don't drink coffee for a few days the headaches will subside. But it doesn't actually enhance productivity beyond placebo.

[kortex]

It does objectively improve productivity though, beyond offsetting withdrawal.

https://www.sciencedirect.com/science/article/pii/S014976341...

[Balgair]

I'd be interested to know what the controls were for those studies. Were the participants already addicted or was the 30mg+ dosing done on non-addicted people? It's a lot of studies to pour through.

Also, that is a lot of metrics!

And it seems that the athletic performance increase to get statistical validation (for any person) is in the grams range. I ... I just can't see any reason to take that much caffeine unless I'm at the Olympics. I'd be jumping out of my skin!

[tehjoker]

I'm not so sure. The stimulation can also self-medicate for people with attention issues. I've tried quitting coffee before for weeks and I get so spacey it is difficult to work on major projects. I try coffee again and suddenly I feel quite capable. Perhaps I didn't quit long enough, but at this point after multiple attempts quitting with similar results, I've just accepted it.

[nchmy]

Because you haven't detoxed long enough. Try 1-2 months - you'll be shocked by the clarity, stability in energy, better sleep etc

[normie3000]

Enjoyable analogy.

> Some choose to spend a lot on appliances that let you brew at home rather than relying on some external provider.

This makes it sound like buying brewed coffee is the budget option. But the real budget option I've seen is to brew at home. Almost any household will have an appliance to boil water. Then add instant coffee.

I don't understand why, but in my experience instant coffee seems to be the baseline even in coffee-producing countries.

[arctic-true]

I think the idea is that there are higher startup costs to brew at home. Even a cheap coffee machine is going to cost more than a cup of coffee at a diner, in the same way that a computer that can run a local LLM is going to cost more than a bunch of API calls to a commercial model. Eventually, those diner coffees add up, but you’re stuck with them if you can’t afford the coffee machine.

[normie3000]

> Even a cheap coffee machine is going to cost more than a cup of coffee at a diner

I think I understood but disagree - the cheapest "coffee machine" is a kettle or cooking pot.

[nchmy]

Cowboy coffee can be excellent. It's almost literally what professional coffee tasters do when they do "cuppings".

A French press just adds fine mesh. You don't even need the glass jar - literally a jar or any other vessel would work

[nchmy]

They're not talking about instant coffee, which is awful. And coffee-producing countries are poor, hence they drink dirty sugar water while the good stuff gets exported.

[glitchcrab]

> But the quality is really no different.

Hard disagree. As someone who is somewhat into the home brewing rabbit hole, I can tell you that the gulf between what I can make at home and what you get in Starbucks is enormous. And I'm no expert in the field by any means.

The rest of your analogy holds up, but not that sentence.

[eudamoniac]

Not the best analogy because coffee has massive quality differences that anyone can tell in a real comparison. Brewing at home is also cheaper and requires nothing more than a $20 French press.

I have personally administered double blind taste tests to four unrelated coffee drinkers who believed their current coffee (Keurig, Nespresso, pre ground) was good. They all thought the actual good coffee was much better in the blind tasting, to the extent that they all permanently stopped drinking their previous coffee. Exclamations of shock occurred.

Coffee is a commodity only out of ignorance. It should be treated as fresh produce of numerous varietals.

Have there been any such "tastings" of LLMs?