Ever since I saw the AI-generated movie The Heist (and wrote about it a few weeks ago) I haven’t been able to stop thinking about it. Partly because I think it’s a pivotal moment for art and film. And partly because the algorithmic gods of Threads and LinkedIn have discovered that I’m hungry for this Heist discussion and have been serving it up to me daily.
Today, another creative unveiled their AI creation: A short film called Lynx that was created entirely using Veo 2.
The debate surrounding both videos - and all of these new AI videos is fascinating. And it comes down to a question of how we describe things, but also a question of morals.
How we describe things - and name things - is important. It shapes the conversation we’re having about these tools.
Traditional creatives, editors, and filmmakers see the term Generative Artificial Intelligence and put the emphasis on the “Artificial Intelligence” rather than “Generative.” They are giving agency to the tool itself, and judging both it and the output as they would judge a creator.
All of the criticism they are throwing at films like The Heist is totally valid: there is no real story. The pacing is terrible. It’s just a collection of shots thrown together. There is no consistency between scenes or even within scenes.
If someone were to create this film by hand and say “hurrah! I’ve made a short - what do you think?” they would likely receive similar, though less scathing criticism. In fact, a short film I made more than 20 years ago would probably be worthy of that same feedback: There is no story, the pacing is bad, the shots seem...cobbled together.
The people actively using and exploring and playing with these tools would agree. In a lot of cases, they likely know that what they are creating isn’t good enough yet. But that it will be soon. As the saying goes, “this is the worst it will ever be.”
CGI changed filmmaking. Digital video changed filmmaking. And in both cases, the quality of the film isn’t dependent on either technology. It’s dependent on the skill of the creator to tell the story they want using that technology.
Last week there was a great post on Reddit about someone using Stable Diffusion to create a scene. The end product was “meh,” but you can see how much work it took to get there. How much finessing and finetuning. And a lot of that is only possible (for now) with the support of a talented artist: someone who knows how to create and how to bring their vision to the screen. It’s not just prompt-engineer.
or these types of creators, the emphasis on Generative Artificial Intelligence is on the Generative. They’re relying on their own intelligence to come up with a story and an idea.
A better term for the various tools, from Midjourney to ChatGPT to Stable Diffusion to Google’s VEO is probably Artificial Image Generation: it still requires imagination, artistic ability, creativity, storytelling, and the understanding of story and creativity.
I sit on the fence between the old guard and the new wave: My career was built on exploring new technology, unpacking it, and helping other people bring it to life (this blog, started in 2006, is a testament to that).
Because of that, I’m excited about the promise of Artificial Image Generation tools. It represents a new of way of creating, and giving the tools of creation to more people.
But I also stand with the traditional crowd of creatives : A film isn’t just a collection of shots thrown together. A film is a story, an idea, emotion, and cuts at the right moment to dial-up or release tension. It’s pacing. And for it to be art, it needs human effort to create it. These tools cut a lot of that out of it.
The question of morality is more important than the question of semantics, though. These AI films wouldn’t be possible if they hadn’t ingested all of the good, human creative work of the last few decades. The corporations that control them stand to profit immensely from this wholesale ripping of art and plan to give nothing back.
Generative AI (and the Pandora’s Box of misinformation, dead internet, and general decline of human culture and creativity that comes with it) is bad.
New tools to facilitate the creation of art in ways that continue to support artists and make new and original artefacts are good.
The line between the two is blurry and wide. The good artistic intentions that have been buried in that in-between space will be feasted on by monolithic organizations creating profit instead of art.