Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

No elephants: Breakthroughs in image generation

oneusefulthing.org

Post reply on HN
Print view View original

311–320 of 373 posts

Re: No elephants: Breakthroughs in image generation

#311

Post by thwarted »

Taek wrote: post #198
thwarted wrote: post #168

Earlier quoted context omitted.

> Its also not clear for example that Studio Ghibli lost by having their art style plastered all over the internet. I went home and watched a Ghibli film that week, as I'm sure many others did as well. Their revenue is probably up quite a bit right now? This sounds like a rewording of "You won't get paid, but this is a great opportunity for you because you'll get exposure".

Exposure has value! The meme around trying to pay artists with exposure is because some people think their "exposure" has meaningful value when they are offering to expose the artist to 100 people, 99 of whom aren't likely even target customers. Studio Ghibli on the other hand had exposure to millions of people (maybe hundreds of millions), and probably >5% of those were potential customers. So yes, being paid in exp…

> Studio Ghibli on the other hand had exposure to millions of people (maybe hundreds of millions), and probably >5% of those were potential customers.

There's a lot of ifs in here. The number of people exposed to has an estimate that covers two orders of magnitude, "maybe". "probably". "greater than". "potential".

In order for this exposure to have more value than the ownership of the original, all of those things need to fall into place. And no one can offer meaningful exposure based on the off-chance that a meme goes viral. All the risk is on the creator, they lose control of their asset and receive a lottery ticket in return.

> So yes, being paid in exposure makes sense, if the exposure is actually worth what the art is worth. But most people offering to pay in exposure are overvaluing their exposure by 100x or more.

Yes, but that's a big "but"; it's difficult to know the value of the "exposure" that is being offered, not to mention if the entity offering it is legit or if it's just a scam because they don't want to pay.

Additionally, the AI companies who are slurping up copyrighted works to train their models are not offering exposure. And the mememaker who happens go viral can't offer it either.

Re: No elephants: Breakthroughs in image generation

#312

Post by __loam »

fc417fc802 wrote: post #306
__loam wrote: post #291

Earlier quoted context omitted.

The people who built this technology needed to use hundreds of millions of images without permission. They regularly speak explicitly about all the jobs they plan to destroy. If you think I'm being hyperbolic then you don't understand the scale of the issue, frankly.

> The people who built this technology needed to use hundreds of millions of images without permission. It remains unclear if they needed permission in the first place. Aside from Meta's stunt with torrents I'm not aware of any legal precedent forbidding me to (internally) do as I please with public content that I scrape. > They regularly speak explicitly about all the jobs they plan to destroy. A fully legal endeavo…

> I'm not aware of any legal precedent forbidding me to (internally) do as I please with public content that I scrape.

Because all the litigation is currently ongoing.

> A fully legal endeavor that is very strongly rewarded by the market.

Yes let's sacrifice all production of cultural artifacts for the market. This is honestly another thing that's being litigated. So far these companies have lost a lot of money on making a product that most consumers seem to actively hate.

Re: No elephants: Breakthroughs in image generation

#313

Post by xmprt »

Vvector wrote: post #258
__loam wrote: post #225

Earlier quoted context omitted.

[flagged]

* You cannot ethically use a tool that was produced by appropriating the labor of millions of people without consent. You are a bad person if you use it. * I disagree. When you publish your work, I can't copy it, but I can do nearly anything else I want to with it. I don't need your consent to learn from your work. I can study hundreds of paintings, learn from them, teaching myself to paint in a similar style. Copyri…

My main concern with AI is that in a capitalist society, wealth is being transferred to companies training these models rather than the artists who have defined an iconic style. There's no doubt that AI is useful and can make many people's lives better, easier, more efficient, however without properly compensating artists who made the training data we're simply widening the wealth gap further.

Re: No elephants: Breakthroughs in image generation

#314

Post by __loam »

fc417fc802 wrote: post #310
__loam wrote: post #279

Earlier quoted context omitted.

> Do you really expect people to blindly adopt your ideological views if you just state them forcefully enough? This is the view of most people outside the industry.

I can't say that the things you're saying match what I've encountered from nontechnical folks lately. Most of them are entirely apathetic about the whole affair while a few are clearly dazzled by the results. The entire thing seems to be a black box that they hold various superstitions about but generally view as something of a parlor trick. The ones that pay attention to the markets appear to believe some very quest…

You're the second person to mention markets to me in this context. Explains a lot, honestly.

Re: No elephants: Breakthroughs in image generation

#315

Post by bongodongobob »

fc417fc802 wrote: post #297
65839747 wrote: post #266

Earlier quoted context omitted.

The production of works that use AI will drop, while the cost to produce higher quality works - works that don't use AI - will remain the same. All we'll get is more AI slop.

Why presuppose that high quality works can't be produced with the help of (near future) AI tooling? Just because you can produce slop with AI doesn't mean that you have to.

Because there are lots of verysmarts here that think they know what's best for industries they have no training or experience in.

Re: No elephants: Breakthroughs in image generation

#316

Post by dudeofea »

fc417fc802 wrote: post #306
__loam wrote: post #291

Earlier quoted context omitted.

The people who built this technology needed to use hundreds of millions of images without permission. They regularly speak explicitly about all the jobs they plan to destroy. If you think I'm being hyperbolic then you don't understand the scale of the issue, frankly.

> The people who built this technology needed to use hundreds of millions of images without permission. It remains unclear if they needed permission in the first place. Aside from Meta's stunt with torrents I'm not aware of any legal precedent forbidding me to (internally) do as I please with public content that I scrape. > They regularly speak explicitly about all the jobs they plan to destroy. A fully legal endeavo…

"Data Laundering": Commercial entities fund (either with money or compute tokens) academic entities which, in turn, create AI models which the commercial entities sell. https://waxy.org/2022/09/ai-data-laundering-how-academic-and...

Re: No elephants: Breakthroughs in image generation

#317

Post by bongodongobob »

flkiwi wrote: post #157
nasmorn wrote: post #149

Earlier quoted context omitted.

Except they didn’t do any work for the exposure. If a marketing agency had come up and executed the Ghiblify everything model as a PR stunt we would call it the most genius creative campaign of the decade

They did though. The studio engaged in tremendous amounts of work and created good will, in addition to their specific creative works. Their visual style is tied up in that good will. Use of the visual style for profit without consent is, at least ethically, misappropriation of another's value. And "You should be pleased I used your creative work because now more people will know about you and you will make a lot of…

You're entirely missing the point. The average western person has never heard of or seen any Ghibli movies. GPT use is heavily skewed by nerdy types. The average football watching big-bang-theory-is-a-smart-show type person doesn't know what any of this is.

If Ghibli feels like they are getting screwed, they could've taken this opportunity to promote themselves, is the parent's point. If I were in their marketing dept I would have been screaming "guys, non-weeaboo people are seeing our name in the news, let's fucking capitalize!" When has Ghibli ever trended? Set up some screenings or stream Spirited Away on their site for a couple weeks or somethin. If they want to win hearts and minds, that's what you have to do. As of now, it's already out of the MSM news cycle and forgotten.

Re: No elephants: Breakthroughs in image generation

#318

Post by dudeofea »

ldoughty wrote: post #196
__loam wrote: post #170

Earlier quoted context omitted.

If you use this technology, you're actively harming creative labor.

Can you elaborate how there's no possible way to use this technology without actively harming artists? If a classroom of 14 year olds are making a game in their computer science class, and they use AI to make placeholder images... Was a real artist harmed? The teacher certainly cant afford to pay artists to provide content for all the students games, and most students can't afford to hire an artist either.. they perh…

> This technology gives the kids a lot more freedom than a pre-packaged asset library, and can encourage more engagement with the course content, leading to more people interested in creative-employing pursuits.

This is your opinion. I don't see how these statements connect to each other.

You might have heard this: it's helpful to strive to be someone only a few years ahead of you. Similar to this, we give calculators to high-schoolers and not 3rd graders. Wolfram-Alpha is similarly at too high a level for most undergraduate students.

Following this, giving an image generator to kids will kill their creativity in the same way that a tall tree blocks upcoming sprouts from the sun. It will lead to less engagement, to dependence, to consumerism.

Scams beget more scams

Re: No elephants: Breakthroughs in image generation

#319

Post by exodust »

schwartzworld wrote: post #257
nitwit005 wrote: post #215

Earlier quoted context omitted.

> For many, many websites this is going to be good enough. It was largely a solved problem though. Companies did not seem to have an issue with using stock photos. My current company's website is full of them. For business use cases, those galleries were already so extensive before AI image generation, that what you wanted was almost always there. They seemingly looked at people's search queries, and added images to…

AI is mediocre at a lot of things, but it makes a damn fine upgrade from stock photos. This is the art that’s going to get replaced by this tech, shitty low effort stuff. Images where you just need a picture of X because people are expecting a picture. It’s the same with code. I don’t think software engineers will really be replaced, but small web dev agencies have a good reason to be nervous. Why would you pay someo…

3-5 prompts doesn't get you a professional restaurant website.

The key word is professional. A good restaurant website begins with taking good photos of the premises and the food. AI won't come around to your business and take professional photos.

There's a lot of bits and pieces to a website for bookings, content management, menu updates, etc.

HTML templates and themes have been around for a long time. AI can basically spit out those templates and themes, which is great. But there's still a lot to do before you get to www.fancy-dining.com.

Re: No elephants: Breakthroughs in image generation

#320

Post by voidUpdate »

JFingleton wrote: post #201
genewitch wrote: post #177

Earlier quoted context omitted.

I have the Gemini app on my phone and you can interact with it with voice only and I was like oh this is really cool I can use it while I'm driving instead of listening to music. I can never think of anything to talk to an AI about. I run LM local, as well

Have it interview you (as-in a job interview) on your specialisation. Works your interviewer skills. Ask it to teach you a language. DnD works really well (the LLM being the game-master).

DnD does not work really well, I've tried that with LLMs before
Post reply on HN