What the Disney/OpenAI Deal says about a Creative Marketplace Dominated by AI

Disney’s deal with OpenAI last week got a lot of attention because one of America’s biggest most-storied legacy content companies finally made a big bet on AI. There are, however, clues in the deal that put into sharp focus what’s really at stake. This is about much more than a brand giant licensing its IP to an AI company. The deal was announced as Disney’s $1 billion “strategic investment” in OpenAI and licensing of iconic characters, framed by the usual corporate platitudes about partnership and innovation. But here’s the tell: since when does a licensor pay the licensee?

For a century, the stream of compensation for intellectual property ran in one direction. Studios, labels, and publishers owned catalogs while technology companies paid for access to them. Even when those arrangements were, shall we say, marriages of convenience — Spotify and the record labels, Netflix and the studios — the power dynamic of who owned what stayed intact. Content was licensed out while money flowed in.

This deal suspends that natural gravity. Disney isn’t just licensing IP to OpenAI, it’s buying a place inside the machine that will define how IP is absorbed, transformed, and redeployed going forward. Call it an investment if you want, but more accurately it is a redefinition of the creative power matrix. Disney is outsourcing its AI technology needs because it knows it cannot build what it needs alone. At the same time it is paying to have a role in shaping the rules of a system it no longer controls. That matters because Disney is paying for something specific: brand safety. Iconic as it is, Disney is under threat from many sides.

As is its new partner. OpenAI is currently battling copyright lawsuits from all sides, from The New York Times to the Authors Guild. It’s spending wildly on data centers. And it as yet doesn’t have any obvious viable products to show for all that outsized spending. In short, these two need each other.

By partnering with Disney, OpenAI gains a content ally that is a gold standard of intellectual property and one of the fiercest, most aggressive defenders of copyrights. Disney is a reputational shield as much as legal one, which OpenAI desperately needs. In return Disney gets a safe harbor, heading off a Wild West AI landscape that could hallucinate Mickey Mouse into compromising contexts and outputs.

Disney isn’t just trying to mitigate that risk, though. It’s trying to muscle its way into imposing its corporate order on a probabilistic system. Disney is paying to create a “Disney-safe” tier of GPT and Sora calibrated to its legal, reputational, and financial risk profile. A second tier, the open, messy systems can be the domain of the public. The clean, high-fidelity systems will belong to institutions that can afford them.

We’ve seen versions of these marriages between content and tech before. Spotify’s deals with Universal Music Group were not about fair compensation; they were about mutual dependency, with Universal owning a piece of Spotify, thereby allowing it to participate on both sides of the content deals. The difference in the Disney/OpenAI deal is important. Spotify needed licenses to survive. OpenAI needs them to stabilize its legitimacy. When Disney locks its catalog inside a partnership relationship, it doesn’t just protect itself — it shuts competitors out, making it harder for smaller AI firms to compete on quality, safety, and trust. Last week included a dramatic example of this. Disney filed a major copyright lawsuit against Google, accusing it of “massive” copyright theft. Ah, we know this strategy: make a deal with one threat and sue the rest.

Leverage, as evidenced by this deal, is going to become heavily tiered in the AI world.

Large rights-holders with significant content catalogs will negotiate equity, access, and customized safety rails. Smaller creators will have to settle for policies, dashboards, and opt-out links largely dictated to them — if they get anything at all. For a moment we thought AI might flatten hierarchies; instead, it is accelerating stratification. Catalogs become bargaining chips, and scale becomes essential leverage. Everyone else is told the system is too complex, too fast-moving, too inevitable to stop.

And doesn’t this sound familiar? In the digital revolution of the early 2000s, platform scale became everything and Big Tech grew so enormous that it swallowed up everything in its path and set up the rules the rest of us had to abide by. The platform infrastructure that determines what gets seen and by whom has had a devastating effect on traditional arts and culture.

One reason Disney was able to move so quickly is that the deal appears to have bypassed human talent. Writers, performers, and directors, the human labor embedded in Disney’s IP, are not explicitly part of this agreement. The 2023 guild contracts (following extended strikes) established some AI guardrails. But I would argue that those guardrails were essentially meaningless even at the moment they were signed because of misdirected stakes; they were easily breached by advances in the tech.

This deal effectively decouples technology rights from labor rights, and once they separate, they rarely reunite, as we discovered in the rise of the big platforms. By structuring this as a corporate investment rather than a production agreement, Disney avoids talent issues: who owns a performance once it becomes training data, and what rights and ownership survive the transformation?

WGA and SAG-AFTRA unions were quick to protest the deal. And rightly so. This will essentially establish a precedent in which the future of production is negotiated between institutions while labor waits outside.

So this is the short game: make an ally, get access to the tech, buy time, and hope that resulting AI user-created slop doesn’t dilute Disney’s brand. The license is only for three years, so the company will learn a lot.

But there is a long game here that is something else entirely.

For all the protests about the value of human creativity, creative industries have “enhanced” or technically altered human performance for decades. Airbrushing transforms skin and shape in images. Auto-Tune smooths and reverb enriches voices. CGI obliterates the constraints of physics in movies. In this context, AI isn’t a break from enhanced production, it does it with more precision and on an industrial scale. The first killer app here is not synthetic movie stars or replacing humans, but personalization. With high-fidelity AI, Disney can dub its entire library into dozens of languages, preserving vocal tone and emotional nuance while perfectly synching movement. It is a massive efficiency gain that also results in a significantly superior product than subtitles and voice dubbing.

Cool, you say. And the ability to do this appears innocuous, even beneficial. But herein portends something much bigger.

If, looking a little ahead, you can generate new performers, scripts, and worlds that never age, never strike, and never renegotiate, the logic of legacy IP shifts. In this view, Disney’s investment looks less like defending its legacy IP than it does preparing for a transition. Legacy content needs protection now, but the defense could also be a gateway to synthetic content that does not.

Which brings us to the question of abundance, or more rightly, over-abundance. The slop problem is not aesthetic, it is economic. Will Disney-adjacent knockoffs devalue the original? Ubiquitous Disney could be cliche Disney could be generically-meaningless Disney. Meanwhile, a marketplace that sizes up by 100x or 1000x makes it much more difficult for human-produced work to be discovered, let alone get traction.

Disney looks to be making a big gamble with its core value assets. By making Minnie and Mickey raw fodder for anyone with an AI account, then everything can look like Disney and Disney becomes generic, uncreative and boring, endlessly reproducible, infinitely diluted.

So why did Disney jump now? Because, despite its outsized brand awareness, it is under enormous threat, and not just from other entertainment companies. The new landscape will be fought over with user data, scale, content and technology. Google has YouTube and search and data made up of trillions of user behavior interactions, as well as its Gemini AI. Meta has its extensive social graphs. Amazon has cloud infrastructure, a legacy movie studio catalog and retail consumer transaction data. Netflix has platform technology and now Warner’s catalog. Disney had content, but now, through OpenAI, it has access to a machine that has a chance to be reshaping everything else.

It’s not about ownership, but access.

The landscape we’re entering into is still murky, and the systems that determine how it will work will continue to evolve enormously around issues of derivatives, training and transformation. In this brave new world, intellectual property (and copyright) isn’t a shield, it’s currency to be used to buy influence inside the infrastructure that will determine creative power. It’s a fascinating and fundamental shift in how we treat the value of creative work.

Which brings us back to the oddness of the licensor paying the licensee. Disney bought into OpenAI because it understands that the era of legacy studios setting the terms of how content is made and gets to audiences has ended.

That shift portends the reordering of the entire creative landscape, whether movie studio or symphony orchestra or museum or theatre. Oh, and let’s not forget the artists, musicians, actors, designers and writers. It’s not the AI will replace artists. It won’t. But artists depend on the systems of value and distribution and currency to be viable in a marketplace. We are moving from an era where anyone with a camera and an idea could compete to one where viability begins and ends with an access badge that determines whether the system works for you or might not work at all.