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

I think some people have come to that conclusion. Many have joined the big labs.

Whether they’re right or not, we might not get to choose. On some level we’re all just witnesses to this moment in history. It will be whatever it is and if the models keep improving that’s exactly what will happen.

If the s curve flattens out. Things might be different


Well, I guess, better join the people protesting the data centers then. This stuff does not (yet) build itself. Someone has to decide and bring up the capital to do it.

Opposing regulation for selfish reasons is something I can at least understand on a human level. But people who submit to market dynamics as the One Truth, even if they acknowledge themselves that those dynamic are harmful are entering cult territory IMO.


There is a fifteen hundred and thirty sixth dimension beyond that which is known to man.

It is a dimension as vast as space and as timeless as infinity. It is the middle ground between light and shadow, between science and superstition, and it lies between the pit of man's fears and the summit of his knowledge.

This is the dimension of imagination. It is an area which we call the the embedding space.


I don't agree. For example, I think they're very good at writing physics simulators, they're great at writing shaders, they seem to have done a good job formally verifying the signal protocol[1].

That's not all code, to be sure but it's not what I've ever heard talked about as pipe fitting.

But aside from that, I think there are lots of things that made people's lives better and are now irrelevant, and people enjoyed that time. It's gone now, and we'll adapt. But it felt worth capturing the feeling.

[1]https://deepmind.google/blog/introducing-alphaproof-code/


Yes! Its like texture that other decisions can get purchase on. Understand that has been a journey for me


Where would you have started? I've had about five drafts of this piece and kept walking away saying "why does anyone care?"

This was the first shape that seemed to have a kernel where if you related to as a reader, I thought it was a good indication that you'd enjoy the whole piece.


Where were you fifteen years ago! I'd love to hear more about your process if you're interested!


Yeah, I'm just downloading sample packs a drag and dropping the into the timeline in FL Studio, for the most part. Just drag and drop.

Though at that point most of the complexity of the music software becomes unnecessary, and I wonder if there are better (modern) tools for that. I saw a video of someone doing the same thing on a PlayStation 1, apparently MTV made a "game" where you make music in the same way.

https://youtube.com/esckos3O7i4


Yes! I find that whenever I try to steal something, I end up taking some of it's essence, but I end up modifying it so much that it's not really that thing any more anyway.

So I want to be less afraid to do that. Because at the end of the day, if I decide it's too stolen, I can always just not release it.


Agreed... and am realizing, especially with themes/melodies, sometimes it's not always easy to modify them enough to the point that it's unrecognizable, but even then, you could be said to quoting their theme - although, am of course not talking about quoting popular music melodies, no one wants a lawsuit:)


I agree. It's all part of the creative process. I guess I framed it this way because I think it's what my younger self needed to hear because I definitely wouldn't have considered that part the creative work then.

I think there's something here that's a little different from just "refrain from criticism" that I needed to hear which is something like "you can spend a lot of time trying to find the perfect way to do something, or you just do it many different ways and see which you like better" and frankly, I spent so much time noodling, it might have been _less_ overall time if I'd just forced myself to record some very formulaic versions I was averse to, and seen if any of those inspired me to actually get the thing done.

I liked "Instead, it’s about creating one version of a project that will likely either be improved upon in the future or serve as inspiration for something else. The key is to cultivate the commitment and conviction to declare that something is done, for now." from your piece. I think that's what I'm trying to get at.


This piece is great. And it’s totally true that we’re surrounded by walls of slop. I stop reading things all the time when I notice the tinge of AI.

That said, I don’t think that just because this is the way things are, this is this is the way things have to be. Even with our interactions with AI. And I have hope.

Ted Chiang had a great piece in the New Yorker years ago called “Why AI isn’t going to make Art”[1] where he describes something similar. There are 10,000 choices to make in every good pieces of art. AI makes the average choice for every decision and the average choice is derivative and uninteresting. There in I think lies both the problem and the solution. Rather than autocompleting our work, we’d all be so lucky to have a captivated and motivated reader who asks all of the questions this author suggests, so we can get better and better at asking them ourselves and making great art.

[1] https://dmd-program.github.io/dmd-300-master/assets/Chiang-A...


If it were just a problem of making too average choices, it could be fixed by sampling at a higher temperature. But I think there's a bigger problem in that the correct choices depend on the context in a complicated way, and LLMs can't handle arbitrary deep dependency chains, so by default you get the average choice based on superficial cues instead of the average choice based on careful planning. Which is why for coding, chain-of-thought and tool calls are necessary to make the implicit explicit.

Making artistic choices explicit and subjecting them to a planning process with iterative refinement would probably already improve quality a lot (consider image generation models using a prompt rewriting to create more varied and detailed outputs) but to get the most out of it, you'd likely need a model specifically trained for that process. And considering that coding is much more lucrative, it might take a while before we get there.


If you think all of these models aren’t finding important bugs everywhere I think your not being honest with yourself.

In fact I think the opposite is true. The Zcash bug was found with opus 4.6 or something like that. Many worse models currently in the wild might be very capable but not yet industrialized for bug finding.

Another example from today. Cryptography system that’s been around for decades: https://eprint.iacr.org/2026/1630

It’s going to be a bumpy few months


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

Search: