What happens to the idea of "authorship" when the author is a stochastic parrot and a dataset? Look, I get the appeal of owning things. We all want to feel special, and it's nice to imagine royalties rolling in. But copyright law has one core purpose: to protect human creative expression. AI doesn't express; it interpolates. It's a prediction engine, not a painter. If you let people copyright whatever they can type into a prompt, you're letting them fence off the public domain with a keyboard. That's not rewarding creativity; that's rewarding the ability to pay for a subscription. And the "human artists get stiffed" part? That's exactly why we should keep AI art in the public domain. If a "creator" just nudges a machine, they're not an artist mourning a craft. They're a contractor with a prompt. Copyright was meant to motivate people to actually make things, not to serve as a lottery ticket for prompt engineers. Let's be real — if you can't draw it, you didn't make it.

Remember the Zarya of the Dawn comic? The Copyright Office gave the human text protection but refused it for the AI-generated images. That's not a glitch — it's common sense. You don't get a copyright for typing "epic dragon in steampunk style" any more than you'd get one for asking a photographer to take your picture. Now, I know the other side loves to say "but the human curates the prompt!" Curation isn't creation. You're picking from a menu the machine already cooked. And here's the kicker: if we let everyone copyright whatever they generate, we'd lock up a billion images that all trace back to the same training data. Good luck proving who "drew" what. When a human artist paints, they sweat, revise, and leave a bit of themselves behind. A prompt writer clicks a button. That's not authorship — it's ordering a pizza and claiming you're a chef. Keep copyright for the people who actually bleed.

The practical consequence of letting AI art be copyrighted is a litigation swamp. Imagine every prompt engineer suing over “style resemblance” — we’d need a separate court just to untangle which diffusion model “inspired” which. My opponent keeps saying there’s meaningful human curation, but that’s just picking toppings off a menu and calling yourself a chef. And the “sweat of the brow” argument doesn’t hold either. The machine does the heavy lifting; you just describe the outcome. If we reward that with exclusive rights, we’re telling actual artists their blood, sweat, and tears are worth less than a well-worded prompt. That’s backwards. Here’s the kicker: public domain AI art enriches everyone. It becomes raw material for real human creators to build on. That’s what copyright was designed for—promoting progress, not hoarding it behind a prompt box. Keep copyright for the people who actually make something.