Google Flow AI Video Tool Empowers Filmmakers
The Evolving Landscape of AI in Filmmaking: A Creator's Journey
Generative AI filmmaker Dave Clark has explored a multitude of tools for creating his artificially intelligent movies, gaining notoriety for his viral short film, "Battalion." Even in the year since its release, Clark observes that Gen-AI technology has advanced significantly, leading him to believe that past innovations can now be pushed to new heights.
Google Introduces Flow: A New Era for AI Video Generation
On Tuesday, at its I/O event, Google unveiled Flow, a generative-AI video generation tool. This new platform is powered by Google's latest AI model, Veo 3. Dave Clark was among the AI filmmakers who collaborated with Google Labs on Flow's functionality and interface, ensuring it was designed with the needs of filmmakers and creators at its core. Clark expressed considerable impressiveness with the outcomes.
"My mind is still blown by the level of control," Clark shared with IndieWire before Flow's launch. "This is for me as a filmmaker, this is what I've been waiting for. You've put in the 10,000 hours, you struggle through these prompts, and you use these different systems. But all we really hope for is that level of control that I feel like we finally got some sense of it, and it's a long time coming for being so early in it."
Unpacking Flow's Advanced Features for Creators
Flow incorporates camera controls that empower filmmakers to precisely articulate camera movement and visual angles, akin to directing a cinematographer. A standout feature, Scenebuilder, allows users to generate a scene and then extend it while maintaining consistency in character designs and locations. Creators can also upload and merge multiple reference images to match elements like faces, clothing, and locations with hyper-specific detail.
Furthermore, Flow supports more natural language in prompt writing. It understands the visual language of screenwriting and fine-grained technical details such as lens choices, lighting conditions, focal lengths, or film grain. Impressively, Flow can generate sound in real time alongside video, including sound effects, background music, and dialogue.
Flow in Action: Crafting "Freelancers" with AI
Clark and his AI production banner Promise are utilizing Flow to create a new AI-generated short titled "Freelancers." The film follows two adopted brothers who independently become international spies and hitmen. Clark described generating a scene where the protagonists are seated in a diner, complete with an establishing shot featuring a slow dolly-in. He then prompted the model for a dynamic, handheld camera motion, followed by a tracking shot of the characters running.
"And it did it! To see the way that the camera was shaking, it's this exact same way I would probably film it if I was in that restaurant. It was pretty cool, and I haven't seen that level of [control]," Clark remarked.
Pushing Boundaries: Flow's Edge Over Existing AI Video Models
It wasn't long ago that OpenAI unveiled its Sora video model. AI filmmakers at the time were amazed by its "prompt adherence" – the ability to specify an action and then have another action occur involving the same figure within the same sequence. Flow and Veo 3 have demonstrated an even more advanced capability: scenes generated by creators can be dragged and extended on a timeline. This effectively continues the action and can lengthen an individual shot beyond the typical 6 to 8 seconds that models usually generate.
Clark likened this to having the virtual ability to create his own "1917"-style "oner" (a long, continuous take) if desired, with action extending seamlessly from one location to another while the character remains consistent.
Deep Dive: Granular Control and Scene Building with Flow
"Let's say it's a guy driving a car around the Amalfi Coast. With Scenebuilder inside Flow, the director in me was able to block out the rest of the scene," Clark explained. "So maybe it starts off inside the car with the guy driving, then it's POV, then I could switch to a wide shot, and the way Flow works is it actually knows all of the information that was put in, all the prompts, the way the character looks, the way the Amalfi Coast and the road looks, the color of the car. It's able to translate that information into the next cut or scene or edit, if you will, and continue on that narrative."
Clark mentioned getting as specific as instructing the scene to appear as if shot with a 50mm lens, then transitioning up to a 135mm long lens. Flow not only understood these technical specifications but could also integrate them with specific blocking.
An AI-generated image created with Google’s Imagen 4. Courtesy of Google Labs
Behind the Scenes: Google's Collaboration with Filmmakers
Matthieu Kim Lorrain, Creative Lead at Google DeepMind, and Thomas Iljic, Product Lead at Google Labs, discussed with IndieWire how understanding the creative processes of filmmakers like Clark informed Flow's design. They were reportedly impressed by what Clark and others created with what they initially considered to be bare-bones functionality.
"Show and tell is important. They don't want just to type text," Iljic stated. "They want to bring in these elements, these ingredients. They want consistency across shots. They want to have projects to start managing the structure, because it's so many files, wondering how am I going to put this thing together?"
Lorrain added that "customization" is a key term for AI filmmakers. He noted that five different AI filmmakers might have five different workflows, or a single filmmaker might use varied methods for different projects, all of which Flow aims to accommodate.
"Sometimes you need image to video because actually you want to start from a composition. You want to really curate the perfect image and then expand," Lorrain elaborated. "Sometimes it's world-building. And in the case of world-building, that's why we have the ingredients so you can define your characters, find your environment, and they want that. That's the type of control. It's not just control on how they can control the video, but also control on how they can design the creative process for each project."
The Future is Now: Accessibility and the Imperative to Understand AI
Clark's company, Promise, recently announced additional funding from Google's AI Future Fund, as well as Crossbeam Venture Partners, Kivu Ventures, and Saga Ventures. Google is also partnering with two other filmmakers, Henry Daubrez and Junie Lau, for Flow's launch. Flow is now available to subscribers of Google AI Pro and Google AI Ultra plans in the U.S., with plans for expansion to more countries soon.
While Clark is unsure about the specific next frontier in this rapidly evolving technology, he believes that as these tools become more accessible, it is crucial for creators to understand them.
"I always reference the James Camerons and the George Lucases, some of the top filmmakers were always technologists, and I feel like we're going to see that next level, that next school of George Lucas, that are going to be inherently technologists," Clark predicted. "You're going to have to understand technology, whether or not you use it."
