There is No Autonomy: The Myth of the Autonomous Art Generator

February 4, 2025

For as long as humans have made art, we’ve invented tools to help us do it better. From paint brushes to Photoshop, every leap forward has sparked both excitement and skepticism. AI is just the next step in that evolution—except this time, the tool can actively collaborate with us. Think of it like a Photoshop you can speak to, capable of generating new ideas, refining your own concepts, and even surprising us with unexpected results. But just like a paintbrush doesn’t decide what to paint, AI doesn’t replace the artist; it enhances what we can do, helping us work faster, iterate more freely, and explore creative possibilities that might have been out of reach before.

AI is just a new tool in an artist’s toolkit, sitting right alongside pencils, cameras, After Effects, and Blender. The difference is that this tool can respond, adapt, and evolve based on how we use it. It’s not about handing over the creative process to a machine; it’s about giving artists more power, new ways to create, express, and push the boundaries of what’s possible. And honestly? That’s only exciting and awesome.

The Autonomous Art Generator

There’s a confoundingly persistent myth that AI is on the verge of, or already has become an autonomous creative force, capable of replacing human artists altogether. That there is some form of an artistic “Skynet” out there, generating music, books, and movies endlessly with no human interaction.

Let me make this as clear as I possibly can - that is abjectly false. It is science fiction fantasy.

It doesn’t help when people like Ben Affleck say things like this: “AI is a craftsman at best … Craftsman is knowing how to work. Art is knowing when to stop. And I think knowing when to stop is going to be a very difficult thing for AI to learn because it’s taste. And also lack of consistency, lack of controls, lack of quality.”

Why it's a Myth

There is no autonomous AI tool in the world that wakes up and decides to create a masterpiece. AIs such as LLMs and GANs don’t have intent, emotion, or personal experiences to draw from. Every AI-generated artwork is the result of human input, direction, and refinement. The technology itself is more like an incredibly advanced autocomplete, predicting and assembling images, music, or text based on the patterns it has learned from the user’s input. It doesn’t “think” or “feel” in the way humans do, nor does it possess originality in the traditional sense. At best, it’s an ultra-fast pattern matcher, remixing what it has been trained on to generate something new—but only within the parameters we set.

Beyond the technical limitations, there are also practical and economic realities that make AI less of a threat than it may seem. High-quality AI-generated work still requires human oversight, curation, and often extensive editing, which means skilled artists remain essential in the process. Moreover, AI models themselves aren’t free to develop or maintain—training and running them takes immense amounts of data, computing power, and financial investment. Businesses and studios still rely on human creators because AI alone lacks the ability to interpret cultural nuance, emotion, or artistic intent in a meaningful way. Simple concepts that are second nature to humans, such as songs repeating words over and over for musical effect, are completely foreign to current advanced AI models.

Another unavoidable limitation of AI is its need for immense computational power. Every major AI company is in a constant battle to make their models more efficient because running them consumes enormous amounts of energy. No AI company wants users mindlessly generating artwork non-stop, draining resources without constraint. That’s why every creative AI tool has built-in limitations—whether it’s slower speeds, relaxed modes, or strict token caps. It’s not because these companies are being stingy; it’s because AI is expensive and power-hungry. The idea that AI is just sitting around autonomously generating art with no direction other than “replace all human artists!” is laughable. If any AI company found out that their tool was being used in such a way, they would shut it down immediately to prevent unnecessary resource depletion. AI isn't an endless fountain of creativity running in the background—it's a highly controlled, technical, and costly process that requires constant human moderation to remain viable.

If it's Made by a Human = Good
If it's Made using AI = Bad

AI-generated visual effects are often criticized as being “synthetic” or “artificial,” while traditional CGI and VFX are praised as more authentic simply because they are created by human artists. But this distinction is entirely backwards. As a working VFX artist, I can attest that traditional VFX frequently rely on completely fabricated elements—stock assets, hand-sculpted 3D models, and generic motion capture data—all stitched together to create a digital approximation of a character or scene. These tools don’t attempt to reconstruct reality; they create an illusion of it. AI, on the other hand, takes actual images, audio recordings, and visual data of a real person—a LOT of them—and processes them to generate something that is statistically and structurally authentic.

When an AI-enhanced effect reconstructs an actor’s face using millions of real images of that same actor, it is fundamentally a more faithful representation than a manually rigged 3D model that was sculpted from scratch by a dude in a hoodie. Yet, AI-generated effects are seen as “uncanny” or “unnatural” because no “dudes in hoodies” were involved. The irony is that audiences have been trained to accept completely synthetic, human-made VFX as "realistic," while rejecting AI-enhanced visuals that are, in many cases, closer to reality than anything a human could fabricate.

People Who Hate AI Have No Idea How it Works

A major reason for this misconception is that not most—but the mass majority of people simply don’t understand how AI art generation actually works. The technology behind it is vastly different from anything that has come before, making it difficult to intuitively grasp. Take Generative Adversarial Networks (GANs) as an example—when AI creates an image, it isn’t cutting and pasting fragments from existing artwork like a digital collage. It starts from visual noise and, through training, learns how to generate entirely new images based on patterns it has detected.

This misunderstanding is why some people react in horror when they see a GAN-generated painting with what looks like an artist’s signature scribbled in the corner. They assume AI is mindlessly stealing artwork and repurposing it, when in reality, it has merely recognized the common pattern of signatures appearing in that location and re-created it from scratch—just like it reconstructs realistic lighting, anatomy, and perspective. These “fake signatures” are not evidence of theft, but proof of how sophisticated these models have become at detecting and replicating the underlying structures of human art. Yet, critics exploit these images as dystopian horror stories, fueling engagement on social media with outrage rather than acknowledging what’s actually happening: AI is getting better at understanding and recreating visual languages in ways no other tool ever has.

A Revolutionary New Tool

As I have already discussed in my essay “AI Music Has Changed My Life,” AI has completely transformed the way I create music, making the process more intuitive, fun, and creatively fulfilling than ever before. Recently, I’ve written and produced over 100 songs using AI, a pace that would have been impossible with traditional recording methods. As a lifelong musician, I’ve always loved songwriting but struggled with the tedious technical aspects of music production—EQ, compression, mixing—it all felt like an obstacle between my ideas and the final product. And I went to sound engineering college!

But AI has removed that friction entirely, allowing me to focus purely on creating. Now, when I imagine a song, I don’t have to spend hours setting up equipment, tracking parts, and troubleshooting audio issues. Instead, I can enter my lyrics, describe my own genre, and generate music instantly. If I need a different vocal style, a harmonica solo, or a double-bass breakdown, AI makes it possible without requiring me to hire session musicians or struggle through instruments I don't play. And as an artist with some distinct talent limitations, most of the time, AI can bring to life my ideas better than my own imagination or ability. What once seemed like an impossible dream—being able to create and release music effortlessly—has become my reality thanks to AI.

Beyond that, AI has even pushed me to become a better musician. Hearing AI-generated melodies and song structures outside of my usual instincts has introduced me to new chord progressions and styles, challenging me to grow as a songwriter. AI hasn’t replaced my creativity—it’s amplified it, making music more accessible, collaborative, and, most importantly, fun.

Conclusion

What makes AI so exciting isn’t just its speed, power, and efficiency—it’s the sheer joy of discovery. There’s something thrilling about explaining your silly idea to a private, non-judgmental machine and seeing an idea take shape in ways you didn’t fully anticipate. It’s like having a creative assistant that doesn’t get tired, second-guess itself, argue with you, or run out of ideas. Whether it’s generating concept art, experimenting with color palettes, helping with storyboarding, or—literally—anything you can imagine—AI offers artists a way to break out of creative ruts and try things and create work they might never have considered. It’s not replacing human intuition—it’s about expanding it.

So the next time you hear an actor or musician or artist talk about how “Robots Cannot Reflect the Human Condition…” or “AI can write you an excellent imitative verse that sounds Elizabethan. It cannot write you Shakespeare…” or “I just don’t believe that a disembodied mind that’s just regurgitating what other embodied minds have said will ever have something that’s going to move an audience” – remember, there are no actor robots, no Shakespearean bots, or disembodied minds even out there.

It’s just nerds on their computers like me who like to make fun art using advanced tech.

Everybody needs to chill.