Hacker News Reader

> First, labor laws isn’t the answer to technology.

Laws have always been created to deal with new technologies and their fallout. Labour laws are no different.

You should read about the musicians strike of 1942 [1]. When the record player was invented some people thought 'why pay a singer for more than one performance if technology allows you to record them and sell infinite copies'? No one would try that today with Taylor Swift for example.

[1] https://jacobin.com/2022/03/1940s-musicians-strike-american-...

No law was created in your link, except ones that made it impossible to create laws like the one you're implying was created.

By "try that with Taylor Swift", you mean I can trivially torrent a copy of The Eras Tour?

What's the market for a talented saxophone player in a major metropolitan area? It's not zero, but thanks to Spotify, it's not as big as before the advent of recorded music. Recorded music didn't just up the game, it completely changed it. Now you have ultra mega stars that make economic impacts to an area when she tours, but the market for a talented saxophone player in a major metropolitan area just isn't what it once was. I'm not arguing for labor laws, but we'll just have to see where this goes.

Er, Taylor Swift?

Your analogy is wrong, Spotify, or really the phonograph, didn't put musicians out of work. The television did, because once that came along "just" playing well wasn't enough to be a star, on camera presence became more important.

I think the automobile put the saxophonist underemployed. As people moved away from the city core in the suburbs live music venues suffered. The radio, recorded audio,and TV cemented it, but people would always frequent live music if the venues existed. And they do - where there are enough fans to support them. This is why we have a lot of live music places in NYC, etc, even with Spotify and instant gratification.

But I also don’t think the life of the musician is all that dire, at least compared to the portrait artist and tunnel digger. There are plenty of jazz clubs in any major city, and live music that pays enough to at least supplement a day job is a thing. Even at the height of live music your average musician would have needed a day job to get by.

I would argue the biggest issue for talented musicians is economic inequality and the rapid increase in the cost of living. If you want to do music as a main job you need to survive on your partner's income or have generational wealth. There's just no room for someone who is talented to keep themselves afloat playing gigs without having 10 roommates and a day job.

The biggest issue for everyone is economic inequality. You could make the same point about an increasing number of mainstream jobs.

The underlying question is: what are the values of the culture? Predatory selfishness is thrilling [1] for the winners, but ultimately self-defeating - literally suicidal from a sustainability POV, physically and socially.

The neoliberal definition of "freedom" is exactly synonymous with predatory selfishness, so everything else can only suffer.

But it's not everyone's definition of freedom. Plenty of people have other motivations, and this economic model aggressively disenfranchises their values.

[1] In a surprisingly unsatisfactory way.

>Second, if you’ve done much with generative imagery you know the raw experience is difficult to control to a specific visual outcome. Fine tuning, in painting, image to image, control nets, all these things help. But it’s easy to make something stunning, but difficult to make your vision come out from the machine. It’s a tool, like any other. It’ll require rethinking what’s better to do with generative ai and what’s better to shoot directly. But direction, cinematography, and acting are about extreme control to achieve a specific outcome. Generative AI doesn’t help that. However what I’ve seen it really do well is take imagery and add detail and nuance, interpolate scenes, and other feats that will likely make post production much more powerful, requiring fewer reshoots, etc. Creating CGI will be simpler, especially bootstrapping, and l will bet it’ll be extremely heavily used preproduction to create visual story boarding.

The problem is not that there will be 0 humans needed. The problem is that something that used to take 100 people, might be done with 5. And that 95 is still creative jobs, jobs that require education, skill, grit.

You make a living either by hunting and gathering, or creating value for others and exchanging that value for things you need through a generic value token of money. The bar for being able to generate value that can sustain you (let alone a family) was already increasing day by day. But with the sudden spike of generative AI, the bar shot up faster than any time in known human history. With things as they stand right now, the amount of education and experience you need to generate value for others is going off the charts. It will be unfathomable in a few years. Yes, humans will be needed, but not as many to provide what we need. Yet those people will not simply stop existing. What is in the cards for them? That is the problem.

My hope is frankly this nudges us into a post scarcity world where turning the crank to eat isn’t so damn meaningful. A WPA / CCC style program might be useful, tapping people to enrich the world rather than press buttons for food pellets. But these are fantasies and unmoored in the present. We will see.

Given the current state of the world, I don't see this ending in an optimistic lens. At least not in the US.

That's exactly what they do with cameras, you aren't allowed them, and in venues where you're legally allowed to have them thru confiscate it. You can't take a photo of the Eiffel tower either.

I disagree, we are clearly in the end phase. Generative AI (not necessarily LLM) is clearly in a completely different category from the things you describe. Suppose it becomes such that it’s equivalent to a human using all the tools of today, but instantly. What would there be to do? I guess we still need to mine for the natural materials, so there’s that.

Portrait painters and tunnel diggers said the same things. Impressionism, expressionism, modern art, etc weren’t obvious outcomes. The thing AI lacks, and will lack for the foreseeable future as we have a much weaker grasp on its mechanisms, is agency and at the root of that creativity. Most movie making today is a lot of mechanical craftsmanship and a lot of that will probably disappear. The bar for good will be raised dramatically. But the demand for the best won’t disappear, it’ll likely amplify as the stakes get higher.

I will wager you we will see a dramatic realignment in the sorts of work that go into production, but the demand for those skills will be so high and the value so immense that we will see a profusion of new work, better output, and more employment in the arts.

Creatives are already paid very little, so if what you’re saying is true that is not a good sign despite your enthusiasm.

Most of white collar work is very much rooted in the “mechanical” tasks.

And those tasks will emphasize the creative in the future. At one time a human computer made a ton of money. However they’re a novelty today. Mathematics and computer science are a much more creative endeavor using mechanical computers than someone who could tabulate large numbers rapidly in their head.

I do think rote learning is about to seriously take a step back in the world and the future won’t reward the ability to turn the crank nearly as much. The white collar readjustment will be realizing memorizing facts is much less valuable than understanding and synthesizing.

We will have to agree to disagree. The real world is about solving actual problems, not solving arbitrary “creative” tasks. And even if it were, go look at salaries in “creative” professions vs engineering.

"Actual problems" like crypto scams, NFTs, and both mainstream and "new" manipulations of value tokens that don't even exist? Or an industry like health insurance, which doesn't even need to exist?

As for salaries - you can be a hopelessly mediocre engineer, doctor, lawyer, etc, and still earn good money. You can even be a bad professional who is actively incompetent and still earn good money.

The fact that salaries can be hyped so they don't accurately reflect social value is a symptom of the problem, not an explanation for it.

Your examples are strange. Why do you think crypto or NFTs are actual problems I was referring to?

Why shouldn’t health insurance exist? Most lawyers make terrible money. Doctors don’t make a lot of money in most countries.

> It would be like regulating the camera because it put portrait artists out of work, or steam drills because it puts tunnel diggers.

Modern farm equipment put a whole lot of unfree* agricultural laborers "out of work", but most of us consider that to have been a good thing.

*serfs, slaves, indentured servants, etc.

I’m not sure I agree it’s hard to create specific things. The tooling, imo, feels infant, but promising. And it can layer upon itself.

Right now control net can do pretty good character pose control, but you have to make the pose. Well… next step is to generate the pose with a request. Same for faces.

When you're editing in an NLE, you can pick the exact frame where your transition begins. When you're editing in DAW, you can change the volume by tiny fractions of a decibel, for a single frequency. Just about every piece of mainstream media (e.g., excluding social media) has had every single frame, and every second of sound, polished to absolute perfection (or at least one sound engineer or editors perception of it). How is AI going to attain the level of polish (via control) we expect from mainstream media?

I’m not clear what you mean.

If you’re imagining that AI tools mean it has to be a straight text -> output model that’s not correct. They are very much iterative transformations.

There's an enormous golf between just being iterative and control down to the last decibel/frame. Professional creative software is design around this fine-grained control.

Ok but what’s your point? This is not outside the capabilities of AI enabled tooling. You can adjust whatever you want, at whatever granularity you want.

The goal is of course to make this unnecessary, because adjusting a single element doesn’t benefit much from ai. But there’s nothing say you can’t?

"Not outside the capabilities" isn't a strong argument, we want amazing, ten times better than the current tools, e.g., to use the language typical when describing what it takes to beat an incumbent tool.

The point is to look at the patterns in how creative work is performed, it's generally about fine-grained control. E.g., the commonality across a musician, film editor, animator, music producer, photographer, painter, etc... if you're making the claim that suddenly AI is going to replace this need for fine-grained control then I'd put the burden of proof on the supporter of the AI tool to describe why the fine-grained control suddenly won't be important.

I’m not making that claim. I’m claiming that AI tools can do fine grained too.

I thought that's what you were saying here? "The goal is of course to make this unnecessary, because adjusting a single element doesn’t benefit much from ai."

(And re AI also doing fine-grained yeah addressed that above: Just showing up isn't impressive, it needs to be absolutely amazing at fine-grained control to be a real creative tool in my opinion.)

Ugh. Ok

1) you can go as fine grained as you want. This is not a limitation.

2) going fine grained beyond a certain point is stupid. You don’t edit individual pixels in high resolution art. The tools allow it, but it is not a part of practical workflows.

3) the goal is of course to make fine grained adjustments not needed, because the fine grained details are well executed with high level executions. Same as experienced in (2).

If you still think this is worth debating I’d request a specific example of a practical detail that you think is fine grained and not servable by AI.

A frame of a film is just an image and is easily handled. A decibel is not a thing.

This of course all comes down to a philosophy of creativity and art. In my philosophy of creativity and art, fine-grained details is where the art gets made. You can disagree with that, but I'd point to as evidence the design of the software and other tools that creators use are designed to facilitate fine-grained edits. So if you think that fine-grained edits aren't where the creativity and art comes from, then I'd ask why do you think all the tools are designed to offer that level of control? (E.g., DAWs, NLEs, 3D modeling/rendering, photo editing, vector editing, etc...).

There are also "coarse" tools, and a market for them, e.g., iMovie and Canva, and I'd put AI tools in that same category of course. There's nothing new about coarse tools, but those are general for a different audience, and for a different production level.

I.e., another way of saying this is that important thing is not "high-level of execution", Steve Albini didn't become a sought after producer by being technically great. In high-level creation, technical excellence is usually table stakes. He became sought after by having a particular vision for his music production, and that vision was executed by having fine-grained control over every single audio element of what he was working on.

You can believe AI will negate the need for all this, but my point is simply that would be a mighty large coup to how high-level creation works today, ergo extraordinary claims require extraordinary evidence.

Blah blah blah. Please point to a specific example. Ideally for art or video so we have ai reference points