AI, Copyright, and the Content Wars: Legal Challenges and Corporate Shifts
In 2025, the world witnessed an explosive leap in Artificial Intelligence (AI) capabilities, especially in generating content. This rapid evolution has sparked not only excitement but also an ongoing legal and ethical debate, as content creators and companies whose core assets rely on copyright engage in legal battles with the tech giants powering these AI models.
On December 22, Pulitzer Prize-winning journalist and Bad Blood author, John Carreyrou, joined forces with several other writers and journalists to file a lawsuit against six major AI firms—OpenAI, Google, Meta, xAI, Anthropic, and Perplexity. The suit accuses these companies of illegally using their copyrighted works to train AI models. This action follows in the footsteps of U.S. record labels and Hollywood studios, who have also filed similar lawsuits against these tech titans.
The phrase “If you can’t beat ’em, join ’em” encapsulates a prevailing strategy among industries grappling with the rise of AI. This growing cooperation has led to partnerships that allow content owners to reap some of the benefits from the tech giants. A case in point: on December 11, Disney—the largest film production and copyright company in the U.S.—struck a deal with OpenAI, investing $1 billion in exchange for equity.
“No generation of humanity can stop technological progress, and we certainly don’t intend to,” Disney CEO Bob Iger declared. “Even if technology disrupts our existing business models, we must adapt.”
Yet, the very same day, Disney sent a legal notice to Google, alleging significant copyright infringement by its AI models. Additionally, Disney, Universal Pictures, and Warner Bros. Discovery are currently embroiled in ongoing litigation with AI image generation platforms like Midjourney and MiniMax.
At the heart of this legal drama lies the enduring concept of copyright law, which has been protecting creators for over three centuries. The fundamental principle is simple: “permission is required for derivative works or reproduction.” But as technology advances, especially with AI, new questions arise. Who owns the rights to a song generated by AI? If an AI-created design is turned into a product, does that count as infringement?
While U.S. content companies have long fought to uphold traditional copyright laws, the situation is different in China. In the AI era, Chinese tech giants have mostly worked to establish partnerships with content owners or created their own revenue-sharing models. In this way, the ownership of copyright often takes a backseat as these companies focus on using content as data for training their models.
This trend is mirrored by the massive increase in AI-generated content. According to mobile data firm Sensor Tower, global downloads of AI-integrated apps hit 7.5 billion in the first half of 2025—a 52% increase from the previous year. The growth of platforms like ChatGPT, Kuaishou, and Google’s Nano Banana reflects this surge, which has also raised concerns about the quality of AI-generated content, with critics pointing to an influx of low-quality material that’s easy to replicate but difficult to regulate.
One cultural shift that has come with the rise of AI-generated content is the growing acceptance of what Merriam-Webster dubbed its “Word of the Year”—”slop.” Once referring to leftover food, “slop” now refers to low-quality, mass-produced online content generated by AI. In their definition, Merriam-Webster noted a curious irony: while people typically scorn “slop,” they continue to engage with it.
The rapid growth of AI-generated videos—now commonplace on news platforms and short-video services—has added another layer of complexity. These videos, at times indistinguishable from real footage, are forcing platforms to develop increasingly sophisticated AI detection methods. The challenge, however, is clear: in the battle of AI vs. AI, it’s unclear who will come out on top.
From Litigation to Collaboration
The entertainment industry’s fight against AI has seen several high-profile lawsuits, but it has also led to some surprising partnerships. Take, for instance, Universal Music Group’s (UMG) strategic deal with Udio, an AI music platform, which followed a year-long legal battle. In June 2024, UMG, along with Sony Music and Warner Music, filed suits accusing Udio and Suno of using their copyrighted music to train AI models.
“Music is driven by fans’ desire to interact with artists,” said a UMG executive. Interestingly, research showed that 50% of U.S. music consumers are intrigued by “AI music” but are not keen on the idea of “AI singer clones.” This underscores the central role that human artists continue to play, even in an AI-driven future. AI may optimize music recommendations, but it’s the artist that remains the key.
Despite the legal battles, AI music companies like Suno are thriving. Suno, which boasts nearly 100 million users, completed a $250 million Series C funding round in November 2025, reaching a $2.45 billion valuation. The deal with Warner Music—another example of shifting from litigation to collaboration—signals the industry’s attempt to embrace AI, provided that it’s done responsibly.
Meanwhile, OpenAI’s video generation model, Sora 2, has sparked both panic and cooperation. The release of Sora—an AI capable of generating videos featuring likenesses of celebrities—caused an uproar in Hollywood. But just 20 days later, the entertainment industry had agreed to OpenAI’s proposed copyright protection system, allowing artists to opt-in for their likenesses to appear in AI-generated videos.
This shift reflects the growing recognition that AI can be a valuable tool, but only if it respects existing intellectual property rights.
Managing Data or Managing Content?
While much of the copyright dispute in film and music has been settled through licensing, the content industry is still grappling with how to handle AI’s use of training data. In 2025, Disney, Universal Pictures, and Warner Bros. Discovery filed lawsuits against AI platforms Midjourney and MiniMax, accusing them of scraping copyrighted images for training purposes.
MiniMax, which is currently in the process of an IPO, has defended itself, claiming that character images should not be considered independent “works.” However, if the court rules against them, the financial consequences could be steep.
The problem for AI companies is not just legal; it’s logistical. With the internet making it easier than ever to collect massive amounts of data, companies are often using content—often without permission—as fodder for their AI models. This raises the question: How will copyright law evolve to deal with this new reality?
Opinion: The Need for Balance
As we navigate the complexities of AI-generated content, it’s clear that we’re standing at a crossroads. On one side, we have content creators who are rightfully protective of their intellectual property. On the other side, we have tech giants pushing forward with innovation, offering unprecedented opportunities but also potential risks.
The key moving forward will be balance. Content creators need to be fairly compensated, and AI should not be used as a loophole for bypassing copyright protections. At the same time, the AI industry needs to be encouraged to innovate without stifling progress with overly restrictive laws. The solution lies in collaboration, in ensuring that as we embrace new technologies, we don’t sacrifice the very creators who make those technologies possible.
Related Analysis
The Future of AI: AGI, ASI, and How Superintelligent Machines Will Revolutionize Our World