In addition to the possible business threat, forcing OpenAI to identify its use of copyrighted data would expose the company to potential lawsuits. Generative AI systems like ChatGPT and DALL-E are trained using large amounts of data scraped from the web, much of it copyright protected. When companies disclose these data sources it leaves them open to legal challenges. OpenAI rival Stability AI, for example, is currently being sued by stock image maker Getty Images for using its copyrighted data to train its AI image generator.
Aaaaaand there it is. They don’t want to admit how much copyrighted materials they’ve been using.
That’s where we differ in opinion. I create art because it’s what I enjoy doing. It makes me happy. Would I like to profit from it? Sure, and I do, to some extent. However, you are conflating two ideas. Art created for profit is no longer art, it is a product. The definition fundamentally changes.
I’m a writer, a photographer, and a cook. The first two I do for pleasure, the last for profit. If I write something that someone deems worthy to train an AI on, first, great, maybe I’m not as bad as I think I am. Second, though, it doesn’t matter, because when I wrote what I wrote, it was a reflection of something that I personally felt, and I used my own data set of experience to create that art.
The same thing goes for photography, though slightly differently. When I’m walking around with my camera and taking shots, I do it because something has made me feel an emotion that I can capture in a camera lens. I have also done some model shoots, where I am compensated for my time and effort. In those shoots, I search for art in composition and theme because that’s what I’m paid for, but once I finish the shoot, and I give the photographs to the model, what they do with them is their own business. If they use them to train AI, then so be it. The AI might be able to make some 99% similar to what I’ve done, but it won’t have what I had in the moment. It won’t have the emotional connection to the art that I had.
As far as the third, cooking, goes, I think it’s the most important. When I follow a recipe, I’m doing exactly what the AI does. I use my data set to create something that is a copy of something someone else has done before. Maybe I tweak it here and there, but so does AI. I do this for profit. I feed people, and they pay me. Do I owe the man who created the Caesar Salad every time I sell one? It’s his recipe. I make the dressing from scratch just like he did. I know that’s not a perfect example, but I’m sure you can see the point I’m making.
So, when it comes to Art v. Product, there are two different sides to it, and both have different answers. If you are worried about AI copying “art” then don’t. It can’t. Art is something that can only be created by the artist in the moment, and may be replicated, but can never truly be copied, in the same way that taking a photo of the Mona Lisa doesn’t make me DaVinci. However, if it’s a product, then we are talking about capitalism, and here we can see that there is no argument against AI, because it is only doing what we have been doing for forever. McDonalds may as well be the AI of fast food burgers. Pizza Hut the AI of pizza. Taco Bell the AI of TexMex. Capitalism is about finding faster, cheaper ways of producing products that people want. Supply and demand. If someone is creating a product, and their product can be manufactured faster, and cheaper, by the competition, then the onus is on the original creator to find a way to stand out from the competition, or lose their marketshare to the competitor. We can’t hamper AI just because some busker is having a hard time selling his spray paint on bowl planet scape art. If you mass produce for the sake of profit, you can’t complain when someone out-mass produces you, AI or human. That’s the way of the world.