The creative industry faces an unprecedented challenge: artificial intelligence companies are using millions of copyrighted works to train their models, but the cost of fighting back through the courts has become so expensive that even major corporations can’t afford to pursue every case of infringement.
This reality has come into sharp focus as AI-generated content floods the market, leaving artists, photographers, writers and other creators scrambling to protect their intellectual property. The problem isn’t just about technology—it’s about access to justice when legal fees can reach millions of dollars for a single case.
A stark illustration of these mounting costs comes from Getty Images CEO Craig Peters, who recently revealed that his company has ‘dumped millions’ into just one copyright fight against Stability AI. Speaking to CNBC, Peters described the legal expenses as ‘extraordinarily expensive’, adding: ‘Even for a company like Getty Images, we can’t pursue all the infringements that happen in one week.’
No ads. No tracking.
We don’t run ads or share your data. If you value independent content and real privacy, support us by sharing.
Read More
Getty’s lawsuit against Stability AI, filed in 2023, centres on allegations that the AI company used ‘more than 12 million photographs from Getty Images’ collection, along with the associated captions and metadata, without permission from or compensation to Getty Images’. The case highlights a fundamental tension in the AI industry: companies argue that paying licensing fees would slow innovation and put them at a disadvantage compared to Chinese rivals who aren’t bound by US copyright law.
The Scale of the Problem
Copyright infringement in the digital age has created what legal experts describe as a ‘legal tsunami’ across creative industries. Over 30 copyright lawsuits are currently pending against AI developers in US courts, representing cases from authors, musicians, visual artists and news outlets. These cases involve some of the biggest names in tech, including OpenAI, Anthropic and Meta Platforms.
The Authors Guild has been particularly vocal about the costs involved, stating recently: ‘Licensing the copyrighted materials to train their LLMs may be expensive—and indeed it should be given the enormous part of the value of any LLM that is attributable to professionally created texts.’
For individual artists and smaller creative companies, the financial barriers are even more daunting. The case of Modern Dog Design provides a sobering example: the Seattle-based design studio was recently forced to sell their studio to cover legal costs associated with a copyright battle, demonstrating how these fights can destroy the very creators they’re meant to protect.
The Fair Use Defence
AI companies have largely relied on the ‘fair use’ defence , arguing that training their models on copyrighted material falls under exceptions to copyright law. This defence allows limited use of copyrighted material for purposes such as criticism, comment, news reporting, teaching or research.
However, courts have yet to definitively rule on whether AI training constitutes fair use. A recent decision in Thomson Reuters Enterprise Centre GMBH v. ROSS Intelligence Inc. marked the first major AI copyright ruling of 2025, though the case focused on a legal research tool rather than generative AI.
Judge Colleen McMahon highlighted the complexity of these cases in November when she dismissed a lawsuit against OpenAI filed by Raw Story and AlterNet, noting that copyright law wasn’t designed to address the novel challenges posed by AI technology.
The Innovation Argument
Tech companies maintain that strict copyright enforcement would stifle innovation in AI development. They point to competitors in China who continue to scrape web content freely, arguing that licensing requirements would put US companies at a competitive disadvantage.
Peters of Getty Images pushed back against this narrative, calling it a ‘world of rhetoric’ that threatens to undermine creators’ rights. In a submission to the Trump administration’s AI Action Plan, Getty argued: ‘US copyright laws are not obstructing the path to continued AI progress. Instead, US copyright laws are a path to sustainable AI and a path that broadens society’s participation in AI’s economic benefits.’
The debate reflects broader concerns about safety and transparency in AI development, as leading tech firms face increasing scrutiny over their practices.
Alternative Approaches
Recognising the prohibitive costs of traditional litigation, some organisations are exploring alternative enforcement mechanisms. The Copyright Alternative in Small-Claims Enforcement Act (CASE Act), enacted in 2020, provides a more affordable option for small copyright claims through a dedicated tribunal system.
Meanwhile, Getty has attempted to demonstrate that AI can be developed responsibly through its own ‘socially responsible’ image generator, which pays artists for their contributions. This approach represents one possible path forward: licensing agreements that compensate creators while enabling AI development.
Several tech companies have begun negotiating licensing deals with content creators. News Corp, which owns The Wall Street Journal, has struck agreements with OpenAI and other AI companies for access to its content. These deals provide a precedent for how the industry might evolve towards more collaborative arrangements.
The Road Ahead
2025 is shaping up to be a pivotal year for AI copyright law. Multiple high-stakes cases are moving towards trial, including the visual artists’ lawsuit Andersen v. Stability AI, which was among the first copyright cases filed against AI companies in early 2023.
Legal experts predict that the outcomes of these cases will establish crucial precedents for how intellectual property rights are balanced against technological innovation. For creators, the stakes couldn’t be higher: the decisions made in courtrooms this year will likely determine whether they can financially protect their work in the age of AI.
The broader challenge extends beyond individual lawsuits. As Peters noted, the current system allows only the largest companies to pursue copyright protection, creating what amounts to a two-tier justice system. Smaller creators—often those most vulnerable to AI-powered infringement—are effectively priced out of legal protection.
The situation has contributed to growing concerns about AI’s broader societal impact, as questions mount about the technology’s effects on creative industries and workers’ livelihoods.
Finding Sustainable Solutions
The resolution of this tension between innovation and creator rights will likely require more than court decisions. It may demand new legislation, industry standards or technological solutions that can protect creators’ interests while enabling continued AI development.
Some industry observers worry that the current trajectory mirrors concerns about AI investment sustainability, where massive spending hasn’t yet delivered proportional returns. The legal costs of copyright battles could further complicate the economic equation for both AI companies and creative industries.
For now, the creative industry finds itself in an uncomfortable position: watching AI companies profit from their work while being unable to afford the legal battles necessary to stop it. The question isn’t just about copyright law—it’s about whether the justice system can adapt quickly enough to protect creators in the digital age.
Peters summed up the challenge facing Getty and countless other rights holders: ‘We can’t pursue it because the courts are just prohibitively expensive. We are spending millions and millions of dollars in one court case.’ Until this fundamental access-to-justice problem is addressed, the copyright crisis in the creative industries will likely continue to worsen, leaving many creators without meaningful legal recourse against AI-powered infringement.
