From Creation to Curation: How AI Is Reshaping Hollywood's Fandom Economy

It adds value by finding and extending the intellectual property that brands already own

The biggest idea to emerge from a recent gathering of entertainment and technology leaders was also one of the most practical: The next chapter of artificial intelligence in Hollywood may be less about creation than curation.

As generative tools make it easier to produce more material, abundance no longer equals value. The more urgent opportunity is to use AI to surface the strongest concepts and understand what already exists inside vast catalogs. This will help teams decide what deserves to reach an audience.

“It is less about necessarily just creation. It is about curation,” said Vamsi Duvvuri, EY’s Americas AI leader for technology, media, entertainment and telecommunications, during Variety’s Entertainment and Technology Summit. In practice, that means using AI to learn faster, not simply to increase output.

Entertainment development traditionally moves from an idea through character and story development, visualization, production and release. AI can compress the earliest stages, allowing teams to build rough representations and test variations before committing substantial time and capital. For experiential projects, where an attraction may take 12 to 24 months to open, that flexibility matters. The technology available at launch may be very different from what existed when planning began.

AI can surface gems in classic catalogs

Curation also begins with knowing what a company owns. Shari Reich, NVIDIA’s global head of audio and music developer relations, argued that some of the greatest opportunities sit inside archives that rights holders have not fully organized or explored. “The next thing with AI is going to be the assets that we already have, or the IP that you already have,” Reich said. “You have companies sitting on crown jewels—millions and millions of tracks, millions of videos or studio demos—that they do not even know are literally sitting in a warehouse.”

Research and discovery tools can analyze music by tempo, lyrics, tone, timbre and other characteristics at a speed manual review cannot match. Applied responsibly, those tools can surface overlooked work, introduce catalogs to younger audiences and create new products around assets a company already controls. In music, that could include authorized samples that preserve credit, compensation and artistic integrity.

AI’s commercial promise is not only a flood of new songs, images or scripts. It is the ability to make existing rights more visible and useful. The same logic can expand physical entertainment.

Simon Robinson, president of global experiences and studio operations at Warner Bros. Discovery, described AI as a tool across the experiential value chain, from early concepts through discovery, personalization and follow-up. A process that once required several vendors, multiple pitches and months of work could produce a much wider field of possibilities in a fraction of the time. But speed does not relieve creative leaders of judgment.

“Just because you can do that, it does not mean you should then go on and do everything,” Robinson said. The advantage belongs to teams that can recognize the ideas worth building.

Enhancing the live experience at concerts and theme parks

Inside an attraction, AI could make a programmed character responsive rather than repetitive or turn a familiar Harry Potter Sorting Hat encounter into an experience that changes on a return visit. A child’s choices in a game might even inform what appears during a theme-park visit, bridging a fan’s digital and physical relationship with a story world. These experiences would be generated in the moment but anchored in protected IP and the rules of the fictional universe.

That possibility comes with an important limit. People attend parks, concerts and live events not only to be recognized as individuals, but to feel part of something larger. “As we think about how we use AI to personalize things and develop hyper-personalization within experiences, we need to be careful not to break the power of community,” Robinson said.

A roller coaster produces an individual rush, yet the shared scream is part of the experience. A stadium concert may be emotionally personal, but its force comes from thousands of people participating together. The design challenge is to make a fan feel known without isolating that person from the crowd.

Nancy Hall, U.S. CEO of WPP Media, made a related point about how stories acquire cultural weight. Audiences respond to human behavior, strong emotion and communal experiences, she said, but the format determines how they participate. Short-form video is fast, visual and shareable; long-form work has room to inspire and entertain; live experiences create shared memories. When people can respond, reshape and share a story, “that is what makes stories culture, and not just content.”

That participation also changes the path from awareness to loyalty. The old marketing funnel is now fragmented across social platforms, search, large language models, audio, streaming, creators and communities. Each has a different role: creators can drive discovery, premium content can deliver scale, communities can create belonging and live events can turn interest into memory.

For brands, loyalty is increasingly built through access rather than points: an invitation, a drop, behind-the-scenes material or a one-of-a-kind encounter. Audible has applied that principle to properties such as Dungeon Crawler Carl, bringing its author and narrator together with fans for a live program whose free tickets disappeared within minutes. The draw was not simply more content. It was proximity to the people behind the story and the chance to experience that access with a community.

The company has also placed audiobook material inside a Twitch gaming environment, where it felt credible because listening was already part of the participating creator’s routine. In another campaign, TikTok-style creative moved out of the phone and onto a building in Times Square, combining outdoor media, motion and social behavior. “We made media part of the story,” Hall said. The placement was not a container for the idea; it became part of the idea.

AI can make that approach more responsive by helping brands identify cultural signals, adapt creative across formats and connect digital behavior with physical places. Yet the technology still requires a point of view. Automated optimization can reveal patterns, but it cannot decide which risk is culturally meaningful or which expression feels true to a community.

Using tools to widen the field of vision

Sony Pictures Television chairman Keith Le Goy offered a useful counterweight to the industry’s growing dependence on analytics. Sony’s strategy has been to develop stories, then find the right home for each one rather than force every property through a single distribution system. Its work across Crunchyroll, PlayStation, YouTube, streaming, FAST channels and live extensions shows how curation can follow an audience’s behavior while allowing IP to move beyond its original format.

Data can support those decisions, but it should not replace conviction. “Believe in your gut, believe in your creativity, believe in the audience,” Le Goy said. That may be the most important instruction for leaders adopting AI: Use the tool to widen the field of vision, then make a choice.

Hall framed the mandate for modern marketers just as clearly. “We have to ensure that humans desire us, but that machines recommend us,” she said. Search tools, recommendation systems and large language models increasingly shape what people discover. But earning attention still depends on relevance, credibility and emotion.

The practical role for AI in entertainment is coming into focus: understand the rights a company already owns, test promising ideas earlier, build experiences that can respond to fans and help those fans find their way into a larger community. Success should be measured in profitable growth and richer audience relationships, not in how much material a system can generate.

The companies that lead will be the ones that pair technical capability with restraint. Protecting creators and rights holders, preserving the integrity of beloved worlds and knowing when personalization strengthens a shared experience. Hollywood does not need AI simply to make more. It needs AI to help decide what matters.

Related: The AI Handoff Is Dead. The Feedback Loop Guides the Work