AInterview: Léo Cannone

Where or what was your first point of contact with AI?

My first brush with AI occurred through a website called Gauguin, reminiscent of Paint but with the remarkable ability to transform basic sketches into realistic landscapes using minimal colors. Despite its simplicity, I found it mesmerizing as an artist to witness the transition from rudimentary sketches to photorealistic scenes. Intrigued, I even curated an Instagram account to showcase my creations, juxtaposing initial sketches with their transformed counterparts. Although my focus eventually shifted with the emergence of new technologies, it was this initial experience that introduced me to AI. I delved deeper into the realm of AI later with projects like midjourney, where I explored images and photography, ultimately turning this pictures into motion with the launch of Runway. It was during this phase that I truly immersed myself in the possibilities of AI.

Director, photographer, Graphic and Motion Designer, you are a multi talented Designer oriented in several fields. Could you elaborate on how your background in art and graphic design influenced your approach to using AI in film?

My artistic practice has been built through the use of various mediums, each offering unique possibilities of expression. Painting and drawing for its spontaneity, photography for capturing instants and refining my vision, and filmmaking for bringing stories to life. Each medium complements and enriches the others, and artificial intelligence emerges as a new ally in this creative landscape. It does not replace the practice of other disciplines but integrates harmoniously, offering new dimensions to my creative process. I see image generation by artificial intelligence as a camera directly connected to my imagination.

I used the limitations of AI to develop a creative solution.

How did the limitations of AI, such as the inability to create natural dialogue or human-like characters, influence the aesthetic and narrative style of your short film “Where do grandmas go when they get lost?”

The biggest challenge for me in making this short film was the technical limitations of AI. At the time, it was impossible to achieve coherence in characters, natural dialogue, and consistent facial features. As a result, creating a film with realistic characters and dialogue was excluded, even though it’s something I greatly enjoy.These limitations led me to rethink how to tell my story. I decided to embrace the constraints by creating something similar to a children’s book or fairy tale, with static, front-facing shots. This approach was a direct result of the limitations of the current AI tools, and it inspired a unique creative spark that shaped the film’s aesthetic. The constraints actually enhanced the narrative by fostering a distinctive visual style characterized by fixed, frontal shots. Just as early 3D animation faced challenges with rendering realistic humans, leading to the creation of “Toy Story” with its plastic-looking toy characters, I used the limitations of AI to develop a creative solution. Working within these constraints led to an artistic and narrative style that served the story well, turning technical limitations into an advantage. While I hope future advancements will remove these constraints, they have currently driven creative innovation in my work.

Which tools are you using and why?

For all my current film projects involving AI, I start in a traditional manner, similar to more classical workflows. I draw storyboards and create concept art, either through painting, drawing, or digital art on my iPad. I write scripts and voice-overs in a conventional way. For the AI-specific tools, I use MidJourney to create images that visualize the scenes I have in mind. These images often need further refinement, so I use Photoshop to retouch the backgrounds and characters, ensuring consistency across all scenes. This step is crucial because the scenes generated by AI sometimes lack perfect coherence, even though there is already a good level of consistency. Once the images are polished, I bring them to life using Runway, which adds movement. If further adjustments are needed, I use After Effects to fine-tune the animations. Finally, I edit the film in DaVinci Resolve, following a traditional editing process as I would with any film.

I think AI will never be able to match the unique sensitivity of each artist.

Are there any aspects of human creativity that you believe AI cannot replicate, despite technological advancements?

I think AI will never be able to match the unique sensitivity of each artist. But more than that, even if one day AI reaches a point where it has its own sensitivity, which is possible, I believe the audience won’t connect as deeply with works created by AI. What interests us when we look at a work of art is knowing that there is someone behind it who has lived the emotions they are expressing. That human experience and personal touch are crucial.

What potential developments in AI technology are you most excited about for the future of filmmaking?

In the short term, I’m most excited about the potential for greater control over image creation and improved rendering quality. I look forward to seeing current technical limitations disappear, allowing for more seamless and precise creative processes. This would enable filmmakers to achieve their visions with fewer constraints and higher fidelity, ultimately pushing the boundaries of what can be visually accomplished.

Cool instagram accounts or directors/photographers using AI?

Unfortunately, I’m not very up-to-date on what other creators are doing with AI, but I can recommend my friend Pierre Zandrowich. He has an amazing Instagram account called cinem.ai where he showcases his work using AI in photography and filmmaking. It’s a great example of the creative possibilities AI offers in visual storytelling.