As early as 2022, Ben Affleck observed artificial intelligence slowly making its way into the film industry. As a director, screenwriter, and actor all in one, he saw something others overlooked: the available AI tools simply did not understand how films work. They could not work with shot logic, understand lighting on set, or grasp why one edit makes sense while another does not. Affleck decided to change that.
In March 2026, Netflix announced that it was acquiring his company, InterPositive, and bringing the entire sixteen-member team of engineers, researchers, and creators under its roof. Affleck himself is joining the streaming giant as a senior advisor. The financial terms of the deal were not disclosed.
InterPositive was developed in secret for 4 years
Affleck did not start with vast amounts of capital or an army of developers. He started with a question: what do filmmakers actually need? And then he went on set.
InterPositive operated in complete secrecy from the time it was founded. Affleck and a small team filmed their own dataset on a controlled soundstage that faithfully replicated the conditions of a real film production. He wanted AI to speak the language cinematographers and directors already know. To understand what happens on set, what can go wrong, and how to fix it.
The result was the first model trained in visual logic and editing consistency. The system can handle real production challenges: missing shots, background replacements, or poor lighting. And crucially, it does so in a way that preserves the filmmaker's intent.
What does InterPositive actually do and not do?
Here is what sets InterPositive apart from most of what we imagine when we hear "AI in film." It is not about generating content from nothing. There is no writing prompts into a text box and waiting for a computer to generate an entire scene. Affleck himself explained it clearly: "AI, people generally think of it as creating something from nothing: 'I type something into a computer and I get a movie.' That is not what this is."
The InterPositive system works differently. It takes footage from a specific production, builds an AI model based on it, and the filmmaker can then use that model in post-production. In practice, this means being able to recolor shots, change the lighting, add visual effects, or address continuity issues, all using material that comes directly from the production in question. No outside elements, no generic outputs.
Affleck deliberately built safeguards protecting creative intent into the system. The tools are designed to keep creative decisions in the hands of artists. The technology serves the story, not the other way around.
Netflix and AI
Netflix does not view artificial intelligence as a threat. Quite the opposite. In recent years, the streaming platform has actively used it in the special effects of its original projects and has repeatedly told investors that it is "very well positioned to take advantage of advances in AI." Even so, this acquisition is unusual for Netflix. Historically, the company has been more of a builder than a buyer. It builds its own technology and develops its own tools. The fact that it decided to acquire InterPositive speaks to how unique the toolset created by Affleck's team is.
Elizabeth Stone, Netflix's Chief Technology Officer, summed it up this way: "The InterPositive team is joining Netflix because we share the belief that innovation should empower storytellers, not replace them."
Bela Bajaria, Netflix's Chief Content Officer, added that the new tools will give creators "more options, more control, and more protection for their vision." At the same time, she emphasized that Netflix does not plan to sell InterPositive's technology commercially. It will be available exclusively to the streaming platform's creative partners.
Ben Affleck vs. AI
Why did an actor and director decide to build an AI startup in the first place? Affleck explains it as a sense of responsibility. He saw technology entering the industry he loves and knew that if no one approached it properly, it could do more harm than good. "We need to preserve what makes storytelling human, and that is judgment. The kind that takes decades to develop, experience to refine, and that only people can have," Affleck wrote in his statement.
InterPositive's entire approach is based on one principle: AI as a tool in the hands of the artist, not as a replacement for the artist. The models are deliberately smaller and focused on filmmaking techniques rather than actors' performances. That is no accident. It is a conscious decision. Affleck is not some tech newcomer without context. He is someone who has spent decades on set, knows what happens during filming, and understands why every detail of lighting or editing matters. That is precisely why he was able to create something that truly meets filmmakers' needs.
Hollywood and AI after years of tension
The acquisition of InterPositive comes at a time when the film industry is slowly coming to terms with the idea that AI in content production does not necessarily have to be the enemy. Until recently, concerns about artificial intelligence were one of the main issues behind the Hollywood writers' and actors' strikes. Attitudes are changing. Last year, Disney announced a partnership with OpenAI that will allow characters from Star Wars, Pixar, and Marvel to be used in the Sora AI video generator. Meanwhile, Netflix has quietly integrated generative AI into the special effects of its series and films.
But the acquisition of InterPositive is different. This is technology that was developed directly within the filmmaking environment, for the filmmaking environment. It is not the product of a technology company that decided to conquer Hollywood. It is a product of Hollywood itself. Skeptics, of course, remain. Comments posted under news reports include claims that it is a tool that will ultimately reduce the number of shooting days and limit work for actors and crews. The question of whether the technology will truly protect creators or, on the contrary, threaten them remains open. Only time will tell.
Sources: variety.com and reuters.com
