Google invested roughly $75 million in the independent studio A24 as part of a new research partnership with its DeepMind unit. The deal, announced June 22, 2026, pairs one of the world’s premier AI research labs with a filmmaker-driven company known for titles such as Moonlight, Lady Bird and the viral horror hit Backrooms. But this isn’t a content licensing agreement. It’s something narrower. And potentially more influential.
A24 keeps full ownership of its film library and data. Google gains no rights to train on its movies or scripts. Instead, researchers from both sides will sit together. They will test ideas inside A24’s actual production pipelines. The studio’s directors and producers get early access to emerging models. DeepMind gets unfiltered reactions from artists who have built careers on risk and distinct voice. Feedback flows both ways. That exchange sits at the heart of the arrangement.
The Investment and the Guardrails
Details of the funding emerged quickly. Variety reported the sum aligns with earlier rounds from investors such as Thrive Capital. The capital supports a dedicated research effort inside A24 while giving the studio priority access to DeepMind’s infrastructure and latest prototypes. Eli Collins, VP of product at Google DeepMind, put the intent plainly in the company’s official blog. “By anchoring Google DeepMind’s innovations directly within the creative process, A24 and its filmmakers can help shape new technology in service of their vision and expand their storytelling possibilities,” he wrote. “This hands-on collaboration provides Google DeepMind with invaluable feedback and guidance from leading artists.”
Those words matter. For years A24 has protected its internal methods. It rarely shared exact workflows even with other Hollywood players. Now it opens the door to one of the largest AI organizations on the planet. IndieWire noted the shift: the films themselves are not the asset here. The repeatable process behind them is. A non-exclusive structure lets A24 continue working with other partners while DeepMind iterates on tools that could spread across the industry.
Scott Belsky, A24’s chief operating officer, framed the decision in practical terms. Better AI tools, he suggested, preserve creative control and support risk-taking rather than simply making content cheaper or faster. The contrast with other studio deals is sharp. Disney has clashed with OpenAI in court. Lionsgate has tested Runway. Netflix has worked with smaller vendors. A24 chose a research-first path backed by real money and explicit boundaries.
Demis Hassabis, CEO of Google DeepMind, echoed the mutual learning angle. The partnership lets his team study A24’s singular approach and feed those observations back into model development. No one expects finished movies to emerge from this collaboration anytime soon. Early experiments will likely focus on mundane but time-consuming tasks. Storyboard generation. Shot planning. Rough sound design. Previsualization that respects a director’s taste instead of imposing generic output.
Yet the implications stretch further. If the joint team can produce tools that feel native to A24’s aesthetic, those same systems could adapt for bigger studios or smaller independents. The feedback loop Collins described becomes a template. Artists steer the math. The math improves the artists’ options. Success here would challenge the narrative that AI in Hollywood means replacement. It reframes the technology as a collaborator whose limits are defined, at least in part, by the people who use it daily.
Pushback arrived fast. Just days before the announcement, one of A24’s own directors, Kane Parsons, had called unchecked AI “cultural rot” in an interview. The timing drew notice on social platforms and in trade coverage. A24 later defended the move. Executives told Film Stories the studio wanted “a seat at the table” when decisions about future tools are made. Sitting out the conversation, they argued, would leave filmmakers at the mercy of products built without their input.
That stance reveals a deeper tension. Hollywood has watched generative systems improve at startling speed. Many creators fear homogenization. Others see opportunity in speeding up tedious steps so more time remains for original ideas. DeepMind’s track record with scientific breakthroughs gives the partnership credibility that pure commercial vendors sometimes lack. At the same time, the $75 million check signals Google’s seriousness about entertainment as a proving ground for consumer-facing AI.
Watch the next twelve months. The first public demonstrations will probably look modest. A smarter script breakdown tool. An image generator tuned to avoid the uncanny gloss that plagues current models. Iterative improvements to video previews that match A24’s muted color palettes and precise framing. Each small win will be measured against one standard: does this tool make the final work more distinctive, or does it pull everything toward an average?
Collins closed his blog post with a note of patience. “The specific goals, technical outputs and creative milestones of this initiative will evolve over time,” he said. “As A24 and Google DeepMind’s researchers work side-by-side to test, iterate and build, this partnership aims to expand what is possible in the future of entertainment.”
The sentence feels carefully worded. No grand promises. No timelines. Just a commitment to iteration inside a real production environment. For an industry still arguing over whether AI belongs on set at all, that measured tone may prove the most radical part of the announcement. Real influence rarely arrives with fanfare. It arrives with access, iteration, and the slow accumulation of better decisions. A24 just bought itself a long runway of both.
WebProNews is an iEntry Publication