Adobe and The Walt Disney Company have announced a multi-year strategic alliance that makes Adobe an Official AI Innovation Partner of Disney. The companies announced it on November 9 at Adobe MAX in Miami Beach. Financial terms were not disclosed.
Under the alliance, Disney will work directly with Adobe to build generative AI models on Adobe Firefly Foundry, Adobe's service for custom, brand-specific models. The models will be trained on Disney's character attributes, environments and visual styles. Adobe says they are meant to produce "commercially safe, production-ready" outputs and describes them as "human-centered and artist-led." Disney also gets early access to Adobe's latest AI, creative and media technologies, including Firefly, Premiere and Photoshop, along with dedicated engineering and research support.
Walt Disney Imagineering R&D is already using Firefly Foundry for theme park concept development. According to the announcement, one model turns hand-drawn sketches into fully rendered 2D concept art, and a second custom image model generates on-brand creative assets that stay accurate to Disney franchises. Adobe tools will also support concept work in Disney's parks and cruise development pipelines.
The deal goes beyond parks. Adobe joins StudioLAB at The Walt Disney Studios as an innovation partner to work on internal creative workflows. ESPN Edge, ESPN's innovation center, will explore using AI for fan engagement, content delivery and its archive. The release says co-innovation projects with StudioLAB and ESPN Edge are already underway. The alliance also includes marketing: Adobe will feature Disney's innovation work at its events, including Adobe Summit and Adobe MAX.
Disney owns some of the most valuable characters in entertainment and has been among the most aggressive in protecting them. It is now building generative models on that IP instead of keeping it away from AI. For Adobe, the deal is a high-profile reference customer for Firefly Foundry and its pitch that companies should train private models on content they own rather than use general-purpose image generators.