Artists Form a War Chest: AI Copyright Collective Raises $40M to Sue the Image Generators
The AI Copyright Collective raised $40M to pursue class-action suits against Midjourney and Stability AI over training data, while coordinating with UK and EU regulators on disclosure rules.
The legal pressure on AI image generators just got a funding round. The AI Copyright Collective, an artist advocacy group focused on challenging unauthorized use of copyrighted artwork in AI training datasets, has raised $40 million in Series A funding — money it plans to put directly toward class-action litigation against companies like Midjourney and Stability AI. The announcement, reported by TechCrunch on February 24, 2026, marks one of the most organized and well-capitalized pushes yet by the creative community to hold AI companies legally accountable.
This isn’t the first time artists have tried to take on the AI industry in court. Class-action suits against Stability AI, Midjourney, and DeviantArt were filed as far back as 2023, with artists arguing their copyrighted work was scraped and used to train image generators without consent or compensation. Those cases have been grinding through the courts ever since. What’s different now is the infrastructure: rather than scattered individual suits, the Collective is building a coordinated legal machine, pooling resources across thousands of affected artists into a single organized effort.
Why $40M and Why Now
Series A funding for a legal advocacy group is unusual. Most nonprofit or advocacy structures don’t attract venture-scale capital — which tells you something about how investors are reading this situation. If the lawsuits succeed, the financial exposure for AI image companies could be substantial. The Collective’s model reportedly involves a mix of litigation funding and membership fees from artists, with backers taking a stake in potential settlements or damages. It’s essentially a legal tech play dressed up as a creative rights movement, and there’s nothing wrong with that — the artists whose work was scraped without permission deserve representation that can actually match the legal firepower of a well-funded tech company.
Midjourney and Stability AI have consistently argued that training on publicly available images constitutes fair use under U.S. copyright law. That argument has had a mixed reception in courts so far. U.S. judges have not handed down a clean, definitive ruling on whether AI training constitutes infringement, which is precisely why both sides are still fighting — and why a well-funded collective can shift the balance. More cases mean more opportunities for precedent, and precedent is what this entire legal battle is ultimately about.
The Regulatory Angle
Beyond the courtrooms, the Collective is reportedly coordinating with UK and EU regulators on disclosure requirements for AI training data. This is the softer flank of the strategy: even if you can’t win every lawsuit, forcing companies to publicly disclose what’s in their training sets changes the legal and reputational landscape considerably. The EU AI Act already contains provisions relevant to training data transparency, and UK regulators have been examining similar requirements. If the Collective successfully pushes disclosure rules through regulatory channels, that documentation could then feed directly back into litigation — making it far easier to establish that specific copyrighted works appeared in specific training datasets.
It’s a two-track approach: sue in court, regulate in Brussels and London. The combination is more threatening than either strategy alone.
What Midjourney and Stability AI Are Facing
Neither company has publicly responded to this specific funding round as of publication. Both have been here before — facing legal challenges, arguing fair use, continuing to operate and ship products. Midjourney shipped V7 and continues to grow its user base; Stability AI has had its share of financial turbulence but remains active. The working assumption at both companies appears to be that they can outlast or outmaneuver the legal pressure. A $40M war chest suggests the artists have decided to test that assumption seriously.
Stability AI in particular has faced financial difficulties in recent years, which could make prolonged litigation more painful. Midjourney, privately held and reportedly profitable, is better positioned to fight — but even deep pockets find class-action exposure uncomfortable when the plaintiffs are organized and funded.
What’s Next
Watch for the Collective to file new coordinated suits in the coming months, likely in U.S. federal court, potentially with parallel filings in EU jurisdictions where the legal framework is arguably more favorable to rights holders. The regulatory push in the UK and EU will probably produce formal disclosure guidelines within the next 12 to 18 months — which will either force AI companies to open their training data records or fight that battle on yet another front. Either way, the era of building image generators on scraped internet art without any legal reckoning is running out of runway. The artists finally have a lawyer — a well-funded one.
