Pippa Pays Artists for AI Video — Can It Close the Gap?

A new AI video startup called Pippa is attempting something rare in generative AI: paying artists every time their style is used to generate content. The question is whether royalties alone can heal the rift between creators and AI companies.

What Pippa Is Doing

Pippa, which launched in May 2026, operates a text-to-video platform where users generate short animated clips. Every time a subscriber creates an image or video in a style derived from a human artist's work, that artist gets paid — $0.005 per image and $0.003 per second of video. Artists also receive a cut from a 5% royalty pool funded by subscription revenue ($14.99 to $99.99 per month).

Co-founders Hogan Shrum and Sean Wright frame their model as a corrective to "the bloody history that the AI industry has been built on," according to The Verge. Wright draws a parallel to the music industry's Napster-to-iTunes transition: a disruptive, extractive model replaced by one that compensates creators.

The company has around 800 paying subscribers but only four signed artists, with four more in talks. Artists go through a vetting process before their work trains platform models, and can even submit under pseudonyms — an acknowledgment of how toxic the AI debate has become in creative communities.

The Catch

Despite its ethical framing, Pippa's underlying models carry the same original sin as the rest of the industry: they were trained on art scraped from the internet without consent. The company hopes to eventually switch to in-house artist datasets, but for now relies on open models "with initial training on the broader set of content out there."

Pippa also faces a differentiation problem. Its output doesn't look meaningfully different from other text-to-video tools. The company plans to incorporate ByteDance's Seedance 2.5 model for 30-second fine-tunable video, but competitors will offer the same. Pippa's real moat depends on signing artists — which depends on whether creators ever accept that generative AI can be ethical.

Why It Matters for Builders

Pippa is a live experiment in ethical AI product design. The revenue-share mechanics, pseudonymous participation, and community vetting are patterns worth watching for any AI tool builder who will face the same questions: how do you compensate people whose work makes your product possible, and how do you earn trust when the underlying technology carries a legacy of extraction?

The answer — small per-use payments plus community pools — may not scale. But it's one of the first concrete attempts to move the AI ethics conversation from litigation to product design.