Is paying artists enough to convince them to embrace AI?
Illustrators have spent years sounding the alarm about generative artificial intelligence startups training their models on artists' work without permission. They've pointed out how the practice is tantamount to theft, and in response, many gen AI boosters have argued that it's necessary for the technology's evolution.
The economics of generative AI have always hinged on a quiet assumption: that the raw material—human creative work—can be absorbed at near-zero cost. The current wave of "ethical" AI startups is not a moral awakening. It is a response to a structural problem. Lawsuits, public pressure, and the slow grinding of copyright law have made the old model increasingly expensive, not in dollars, but in risk.
Pippa and its peers represent the next logical step in that calculation. If you cannot train on artists' work without consequence, you pay for it. The royalty model is less a gesture of fairness than a licensing scheme designed to make the inputs legally clean and the outputs commercially defensible. The artists receive a fraction. The startups receive indemnity.
The deeper question is whether the payments are large enough to change behavior. For most illustrators, the answer is no. A few cents per generation does not offset the structural devaluation of their craft. What it does do is create a thin veneer of consent that allows the underlying machinery to keep running.
This is the pattern to watch. Every industry that touches generative AI is moving toward some version of this arrangement—token compensation in exchange for legal cover. The remote and freelance labor market will feel it first, because that is where the flexible, project-based creative work already lives. Platforms that aggregate and pay distributed workers for AI training data are not a side story. They are the supply chain being built in real time.