The Deal, and What Makes It Unusual
Google DeepMind announced a 75 million dollar investment in A24 on June 22, 2026, paired with what the lab calls a first-of-its-kind partnership to co-develop AI filmmaking tools, including AI-generated storyboards. The headline number is in line with Thrive Capital's last A24 round, which tells you the equity stake is conventional venture financing. The partnership attached to it is what makes this notable. This is DeepMind's first known partnership with a full studio, a step up from its earlier collaborations with individual filmmakers such as Darren Aronofsky. Moving from working with a single director to co-developing tools with an entire studio is a meaningful escalation in ambition.
What separates this from a typical AI-meets-Hollywood announcement is the terms. A24 retains full creative control, and crucially, the deal does not grant Google access to A24's content library or data. In an era where the default assumption is that a tech giant's investment comes bundled with a data grab, that carve-out is the headline for anyone paying attention. DeepMind is buying a seat at the table to build tools alongside a respected studio, not a pipeline into that studio's catalog. The structure says as much about the leverage content owners now have as it does about DeepMind's strategy.
Building With Artists, Not Around Them
The stated philosophy behind the deal is collaboration rather than extraction. "We believe the best way to develop tools that empower artists is to work directly with them," said Demis Hassabis, CEO of Google DeepMind. He expanded on the reasoning. "By collaborating with filmmakers and industry leaders like A24 from the beginning, we can build new AI features to support artists in authentic, meaningful storytelling that helps enable their creative vision." The phrase from the beginning is doing real work here, signaling that the tools are meant to be shaped by practitioners rather than imposed on them after the fact.
Whether that philosophy survives contact with commercial reality is the open question, but the framing is a deliberate response to a hostile environment. The creative industries have spent two years fighting AI companies over training data, consent, and credit, and trust is in short supply. A24's Scott Belsky drew the contrast sharply, saying the new tools "won't look anything like the prompted generation type of AI that people feel uncomfortable with." That is a direct rejection of the text-to-video paradigm that has alarmed artists. The pitch is assistive tooling that augments a filmmaker's process, like AI-generated storyboards that speed pre-production, rather than generative systems that threaten to replace the people doing the work.
The Carve-Outs Are the Real Story
Strip away the celebrity studio and the filmmaking glamour, and the structural innovation here is the set of explicit carve-outs. No access to the content library. No access to the data. Full creative control retained by the studio. For enterprise buyers and content owners watching how frontier labs strike domain-specific deals, those terms are the template worth studying. They demonstrate that it is possible to partner with a major AI lab on co-development without surrendering the asset that makes your business valuable, namely your proprietary content and the data it generates.
This matters far beyond Hollywood. Any enterprise negotiating with an AI lab, whether in media, healthcare, finance, or law, faces the same core tension: the lab wants data to improve its models, and the customer needs to protect its IP and meet its compliance obligations. The DeepMind and A24 structure shows one way to resolve that tension, by financing and co-development without data ingestion. Our view is that the carve-outs, not the 75 million dollars, are what corporate development and legal teams should be photocopying. When a lab agrees to build with you while contractually staying out of your data, it sets a precedent that strengthens every content owner's negotiating position.
A Pattern Across the Industry
DeepMind is not moving in isolation. Hollywood's relationship with AI has shifted from outright resistance toward selective, structured engagement, and the comparable moves are piling up. Netflix acquired Ben Affleck's InterPositive, and Amazon MGM stood up a dedicated AI unit. Each represents a different bet on how AI fits into the production pipeline, but together they signal that major content players have concluded the technology is coming whether they shape it or not, and that shaping it from the inside beats fighting it from the outside. DeepMind's choice of an independent, prestige studio like A24 as its partner is a distinctive entry in that pattern.
The strategic logic for the labs is straightforward. Frontier AI companies increasingly recognize that winning regulated and sensitive verticals requires domain-specific co-development, not generic models dropped over the wall. Filmmaking, with its acute IP sensitivities and its empowered creative workforce, is exactly the kind of vertical where trust and tailored tooling matter more than raw model capability. By partnering with A24 on explicit terms, DeepMind is buying credibility and domain insight in a space where both are hard to acquire. The question is whether genuinely useful tools emerge, or whether this becomes another well-intentioned partnership that produces more press than product.
What Could Go Wrong
We would not mistake good terms for guaranteed success. Co-development partnerships between research labs and creative studios are hard to execute, because the two organizations operate on different clocks and incentives. DeepMind ships on research timelines and optimizes for capability; A24 ships films and optimizes for artistic vision and box office. Bridging those cultures to produce tools filmmakers actually adopt, rather than demos that impress at announcements, is the real test. The history of enterprise AI is littered with partnerships that looked elegant on paper and delivered little, and creative tooling is an especially unforgiving domain where artists abandon anything that gets in their way.
There is also the durability of the carve-outs to consider. No data access and full creative control are excellent terms today, but partnerships evolve, and the pressure to feed proprietary content into models grows as the relationship deepens and the commercial stakes rise. Enterprise buyers studying this template should note that initial terms are only as strong as the governance that enforces them over time. The smart lesson is not just to demand carve-outs at signing, but to build in the auditability and renegotiation protections that keep them intact as the partnership matures. A24 appears to have negotiated well. The next several years will reveal whether the terms hold.
The Takeaway for Buyers
For enterprise leaders, the DeepMind and A24 deal is less a Hollywood story than a procurement and data-governance case study. The transferable insight is that you can engage a frontier lab as a co-development partner while contractually protecting your most valuable assets, and that doing so is now a realistic negotiating posture rather than wishful thinking. The labs need credible partners in sensitive verticals badly enough to accept meaningful constraints, which shifts leverage toward content and data owners who know what they are protecting and why.
Our advice is to treat this structure as a starting reference, not a finished answer. Demand the carve-outs DeepMind agreed to here, no library access, no data ingestion, retained control, and then go further on the governance: define audit rights, data-handling specifics, exit terms, and what happens to jointly developed tools if the partnership ends. The 75 million dollars and the A24 brand will get the attention, but the enduring value of this announcement is the proof that a major AI lab will build with you on terms that keep your data yours. That is the precedent worth carrying into your own negotiations.
