Ben Affleck has become an unlikely viral explainer of artificial intelligence. In recent interviews, the actor discussed machine learning, neural networks and visual effects in detail, describing how AI tools can help film crews with particular parts of production.
A long-running interest in computers
Affleck told GQ’s “One More Question” series that he had been interested in computers since he was young. His path into AI, he explained, came as filmmaking moved from analog film to digital workflows, where machine learning had already become part of visual effects work.
He said he can write simple Python scripts and described how image data can be represented numerically for processing. Affleck also explained how neural networks can find patterns in images, including edges and features that help separate a green screen from the scene around it.
That level of detail is part of why the clips caught viewers’ attention. Rather than just naming AI concepts, Affleck paused to explain terms to the interviewer, Zach Baron, and connect them to practical filmmaking work.
Building tools around film production
Affleck said his interest led him to visit companies working on AI, including OpenAI, and to consider how he could participate more directly. His goal, as he described it, was to develop technology that could work alongside artists rather than replace their work.
He said his company created its own dataset for training models on defined tasks. Affleck described raising money, spending about eight months filming with cameras and other equipment, and building data intended for later-stage training of open models. The aim was to connect the training and inference code to specific tasks that could deliver value in production.
Affleck has also discussed fine-tuning open video models by changing their weights and training them toward cinematic standards. In that approach, a film crew can adapt a model for the needs of a movie while keeping its proprietary work. He said AI was used during post-production on his movie “Animals.”
Earlier this year, Affleck sold his AI filmmaking startup to Netflix for a reported $587 million. He said that figure was “not right,” explaining that he did not own the whole company.
Why the dataset matters
Affleck framed the company’s dataset as part of making the technology work with artists and established relationships around likeness. His account points to a practical concern in filmmaking: a model needs to be shaped for the work it is asked to do, while the production’s own material and standards remain important.
That focus helps explain his distinction between AI assistance and AI replacement. If a tool is trained for a discrete production task, it can contribute to a workflow without taking over the entire creative process. Affleck described the technology as something he expects to be additive to the film business.
His concerns are closer to home
Affleck’s comments about AI risks were less focused on science fiction than on everyday effects. He said he worries about his children in school, learned helplessness, responsible use, and what he described as a 30% rise in the number of A’s given out at colleges over the last three years.
He said he does not worry about Skynet or expect AI to take over the movie business in a meaningful way. His position is that the technology can be useful, but its value depends on how people use it and how it is integrated into creative work.
The viral interviews have made Affleck’s technical side more visible. They also offer a glimpse of how he sees AI in filmmaking: as a set of tools that can help with specific production problems, provided artists remain part of the process.