Focus Features released a generative AI documentary suffering from too much access and not enough thought. This pattern is becoming familiar. AI documentaries in 2026 are failing to capture reality.
The Access Problem
AI companies are desperate for positive coverage. They give access, cameras into offices, founder interviews. Sounds like a filmmaker's dream. It's a trap. Eager subjects mean corporate promos with documentary runtimes. Interesting stories are in gaps between what companies say and do, not well-lit conference rooms.
The Complexity Problem
AI is genuinely complex across technical, social, economic, and political dimensions. Single documentaries can't meaningfully cover everything. They try anyway, producing films a mile wide and an inch deep. Best documentaries tell specific stories: one artist disrupted, one company deciding, one community dealing with consequences.
The Timeline Problem
AI moves too fast for documentary production. Started in 2024, released in 2026, covering ancient history. Models changed, companies pivoted, policy shifted. The documentary arrives as historical artifact.
What Would Work
Follow a hospital deciding whether to deploy AI for diagnosis. Show meetings, disagreements, trials, failures. Or follow a researcher who quits a lab over safety. Or follow a town where the largest employer automates. Real stories with real stakes.
These require patient filmmaking, difficult access, willingness to sit with ambiguity. Harder than interviewing seven San Francisco CEOs. But infinitely more valuable.
Why It Matters
How the public understands AI affects policy and governance. Bad documentaries present AI as purely utopian or dystopian. Reality is messy and contextual. People are hungry to understand AI. The filmmaker who cracks this won't have the best CEO access. They'll tell a human story that involves AI.