Google invests $75M in A24 for AI research partnership

Google DeepMind just wrote a $75 million check to the studio behind Everything Everywhere All at Once. The investment marks Google’s first equity stake in a film studio, and it comes packaged with a multi-year AI research partnership that aims to build new tools for movie production and distribution.

The deal, announced on June 22, pairs one of the world’s most advanced AI research labs with one of Hollywood’s most culturally relevant indie studios.

What the partnership actually involves

The core of the agreement is a dedicated AI research lab housed at A24, where DeepMind researchers and A24 creatives will work side by side. The goal is developing AI-driven tools that assist filmmakers across the production and distribution pipeline, not replacing them.

DeepMind VP Eli Collins described the initiative as “first-of-its-kind,” aimed at empowering creatives with advanced technology.

Google does not get access to A24’s content library or data. That’s a significant guardrail. It means DeepMind can’t train its models on A24’s catalog of films, which effectively walls off the partnership from the kind of data-scraping arrangements that have made artists across every creative industry deeply uncomfortable.

A24’s leadership, notably Scott Belsky, has emphasized retaining full artistic control.

Why A24, and why now

A24 has become something of a cultural institution in independent film. The studio’s slate includes critically acclaimed titles like Everything Everywhere All at Once, along with more recent releases such as Backrooms and Marty Supreme.

For A24, the calculus is more straightforward. The studio gets $75 million in equity investment and access to cutting-edge AI research.

What this means for investors

The no-access-to-content-library provision is particularly notable. If this model proves successful, expect other studios and tech firms to adopt similar frameworks, creating a new category of AI-entertainment collaborations that treat creative data as off-limits.

There’s also the question of what “AI tools for production and distribution” actually means in practice. The phrase is broad enough to encompass everything from automated color grading to predictive audience analytics to generative pre-visualization. The value of this partnership will ultimately depend on whether the tools that emerge are genuinely useful to filmmakers or just impressive demos that never leave the lab.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.