concats

Sounds like a solution looking for a problem.

Perhaps. But I can't see a reason why they couldn't still write endless—and theoretically valuable—poems, dissertations, or blog posts, about all things red and the nature of redness itself. I imagine it would certainly take some studying for them, likely interviewing red-seers, or reading books about all things red. But I'm sure they could contribute to the larger red discourse eventually, their unique perspective might even help them draw conclusions the rest of us are blind to.

So perhaps the fact that they "cannot know red" is ultimately irrelevant for an LLM too?

Human entitlement really is the bane of game theory.

There are use cases where even low accuracy could be useful. I can't predict future products, but here are two that are already in place today:

- On the keyboard on iphones some sort of tiny language model suggest what it thinks are the most likely follow up words when writing. You only have to pick a suggested next word if it matches what you were planning on typing.

- Speculative decoding is a technique which utilized smaller models to speed up the inference for bigger models.

I'm sure smart people will invent other future use cases too.

That's a clickbait title.

What they are actually saying: Given one correct quoted sentence, the model has 42% chance of predicting the next sentence correctly.

So, assuming you start with the first sentence and tell it to keep going, it has a 0.42^n odds of staying on track, where n is the n-th sentence.

It seems to me, that if they didn't keep correcting it over and over again with real quotes, it wouldn't even get to the end of the first page without descending into wild fanfiction territory, with errors accumulating and growing as the length of the text progressed.

EDIT: As the article states, for an entire 50 token excerpt to be correct the probability of each output has to be fairly high. So perhaps it would be more accurate to view it as 0.985^n where n is the n-th token. Still the same result long term. Unless every token is correct, it will stray further and further from the correct source.

What about the free open weights models then? And the open source tooling to go with them?

Sure, they are perhaps 6 months behind the closed-source models, and the hardware to run the biggest and best models isn't really consumer-grade yet (How many years could it be before regular people have GPUs with 200+ gigabytes vram? That's merely one order of magnitude away).

But they're already out there. They will only ever get better. And they will never disappear due to the company going out of business or investors raising prices.

I personally only care about the closed sourced proprietary models in so far as they let me get a glimpse of what I'll soon have access to freely and privately on my own machine. Even if all of them went out of business today, LLMs would still have a permanent effect on our future and how I'd be working.

I think one issue is that you won't always be able to invoice those extra 999 hours to your customer. Sometimes you'll still only be able to get paid for 1 hour, depending on the task and contract.

But the llm bill will always invoice you for all the saved work regardless.

The way I see it:

* The world is increasingly ran on computers.

* Software/Computer Engineers are the only people who actually truly know how computers work.

Thus it seems to me highly unlikely that we won't have a job.

What that job entails I do not know. Programming like we do today might not be something that we spend a considerable amount of time doing in the future. Just like most people today don't spend much time handing punched-cards or replacing vacuum tubes. But there will still be other work to do, I don't doubt that.

I won't deny that in a context with perfect information, a future LLM will most likely produce flawless code. I too believe that is inevitable.

However, in real life work situations, that 'perfect information' prerequisite will be a big hurdle I think. Design can depend on any number of vague agreements and lots of domain specific knowledge, things a senior software architect has only learnt because they've been at the company for a long time. It will be very hard for a LLM to take all the correct decisions without that knowledge.

Sure, if you write down a summary of each and every meeting you've attended for the past 12 months, as well as attach your entire company confluence, into the prompt, perhaps then the LLM can design the right architecture. But is that realistic?

More likely I think the human will do the initial design and specification documents, with the aforementioned things in mind, and then the LLM can do the rest of the coding.

Not because it would have been technically impossible for the LLM to do the code design, but because it would have been practically impossible to craft the correct prompt that would have given the desired result from a blank sheet.

Depends on what front end you use. But for text-generation-webui for example, Prompt Caching is simply a checkbox under the Model tab you can select before you click "load model".

Here you go, enjoy!

https://clarkesworldmagazine.com/kritzer_01_15/

Great take. A mixtape can be greater than the sum of its parts!

I believe that a lot of the judgement is also connected to the quality of the works. "Slop", while doubtlessly accurate for today, may be a rather weird description in a couple of year if the rate of progress continues to accelerate like it has.

Although I've already heard people starting to refer to DeviantArt and the like as full of "human slop" so perhaps this is just modern language that's evolving and completely unrelated to AI.

I agree, the story behind the work is part of how we humans view a creation and cannot be dismissed.

I think we're a long way away from 100% algorithmically created content. This far all I've seen is content that is created based on human inputs and ideas. I'm not aware of the Instagram incident you mentioned, but it too seems like the brain child of a human if I'm not mistaken.

There have been trending AI generated videos floating around lately for example. Which I found surprising at first. But they still had a human script writer (prompt writer?), director and a human editor. Someone who had a vision of what they wanted to create and share. My prediction is that this human-directed tool-like usage will be the standard for a long time, so I'm not particularly worried about humans getting removed from the process.

While I have no doubt that individual companies, such as OpenAI for example, will eventually introduce enshittification features, I doubt the industry as a whole can be summarized that easily.

I believe, over all, development will go forward and things will get better. A rising tide lifts all ships, even if some of them decide to be shitty leaking vessels. If nothing else we always have open source software to fall back on when the enshittification of the proprietary models start.

For a practical example: The cars we drive today are a lot better than 100 years ago. A bad future isn't always inevitable.

My theory is that as the quality of these generative tools increase, we'll see the public opinion of them slowly shift. Regardless of philosophy (although discussing it is always fun), it just seems inevitable since there are so many more consumers than producers. And as you say, consumers are the ones that will primarily benefit from this new technology. As a consumer we care primarily (some could argue solely) about our own emotional reaction to the music —or more generally put, art-piece.

In practical terms I also believe that this will give rise to a lot of new consumer behavior, and, as you so aptly puts it "creative consumers" will become normal.

The ability to on-demand create more content to fill out some very narrow niche is a great example ("Today I want 24 hours of non stop Mongolian throat singing neo-industrial Christmas music"). Or maybe to create covers of songs in the voices of your favorite long dead artist. Anything from minor tweak of existing works ("I wish this love song was dedicated specifically to ME", to completely new works (Just look at how much the parody-music genre has grown since Suno and the like first appeared). The possibilities are near endless.

Personally I don't like to gatekeep art.

For example: If someone walks out into the wilderness and encounters a particularly fascinating rock formation or plant, something that was created completely by accident and without a artist or designer, but they find that the sight instills in them strong emotions or deeper thought, I believe they should be allowed to call that art.

Maybe this is just petty linguistics and semantics though, in which case we're drifting away from the topic at hand, and I'm sorry.

Trying to look at the bigger picture for a moment. A lot of the philosophical debate about art I see here, and elsewhere on social media, is often very shallow and can be reduced to:

Does one believe that the value of the art-piece (be is music, paintings, film, or whatever) is created in the mind of the artist, or is it created in the mind of the consumer?

If you believe only in the former, AI art is an oxymoron and pointless. If you believe only the later, you're likely to rejoice at all the explosion of new content and culture we can expect in the coming years.

As far as I can tell though, most regular people think that the truth is somewhere in between these two extremes, where both both the creator and the consumer's thoughts are important in unison. That culture is about where the two meet each other, and help each other grow. But most of the arguments I've seen online seem to ignore or miss this dichotomy of views entirely, which unfortunately reduces the quality of the debate considerably.

"Ten year contract you say?"

"Yes, yes... Satellites stay in orbit for a while. What about it?"

"Looks a bit cramped in there."

"Stop complaining, at least it's a real job, now get in, we're about to launch."

The only similar example I can think of is when, roughly 2400 million years ago (during the Paleoproterozoic iirc), the ancestors of cyanobacteria poisoned their atmosphere by overproducing oxygen which resulted in an extinction event. But that whole process still took somewhere in the order of millions of years to complete I believe.

A lot of people seem to argue that the original intent was for disputes to be handled by the courts. That is, governed by laws outside of the crypto's smart contract implementation. And thus the company is right to seek judicial help and label this act as theft. Alright, sure, on its surface that doesn't seem like an unreasonable position. After all, we all trust the courts right?

But, if this really was agreed upon by most of the parties involved, shouldn't the smart contract have include giving the courts a master-key that allows them to override the blockchain when necessary? Undoing fraudulent transactions and such. Can we really argue that everyone expected disputes to be handled by the courts if this wasn't implemented?

There is no technical reason why it couldn't have been done, as far as I can tell. It would not be great for PR perhaps, since it sort of goes against what a lot of the original crypto enthusiasts believed in: decentralization and protection from future hypothetical tyrannical governments. But at least it wouldn't be half as hypocritical as what we have today.

Of course, if you designed your crypto like this, with a court controlled backdoor, you'd unfortunately have to stop calling the whole thing decentralized. But if that's the intent, and everyone agrees to it, what's the issue?

I'm trying not to pass any value judgement on the "Canadian math prodigy" in this scenario, on the whole I don't care much about the isolated incident, but rather on whoever wrote the smart contract trying to both have their cake and eat it too.

I find crypto really fascinating from a technical and philosophical standpoint, but I'm not too fond of how it's been adapted by society as mostly a sort of get-rich-quick scheme.

> Let’s say I write a book or record an album and there is no copyright. How do I get paid?

A couple of years ago, I would have agreed that this would be a big and complicated issue. But do we all really think this is a problem in 2025? By now we actually have proof that it isn't impossible.

First of all, there is already, today, a plethora of FOSS projects where the developers get regular salaries without relying on royalties. Good experienced people work, make a career, yet the end product is free. We have hard proof that it won't "become a non-viable [...] career". The field might shrink, or at least change, but it won't die entirely. There is still value to be gained, so someone somewhere will pay to have it created.

Furthermore, now a days there are also websites such as Patreon, Kickstarter, and the like that have clearly demonstrated that the same can also be applied to creative endeavors such as art, music, or writing. You just pay people as they work, monthly salaries, rather than afterwards based on royalties. Yes, it comes with other issues, and for many years it was hard to say if those would be crippling. But by now we can actually see, with real world examples, that it's a feasible way to fund creative projects, and the people working on them, even if the digital products were released for free after completion. It turns out that millions of people are willing to pay for these things, to speed up the development presumably, even though they would have gotten the same product for free later, it's no longer a hypothetical.

I also think that the future will show us even more ways to finance creative intellectual work without making things into 'intellectual property' that you have to guard fiercely with copyright as if it was in short supply.

Yes, it would probably become very hard to earn really big money, millions or billions, in the industry if everyone was working for salaries rather than royalties. The distribution would presumably be less top heavy, and more similar to other industries where "going viral" isn't really a thing. But is that such a bad thing? And it's not even a guarantee that this would change, after an artist goes viral their next project would presumably have a lot more fans and investors willing to ensure they continue working, since they loved the first album/film/etc so much.