AI Can Make a Million Images. So What?

Generative AI can render a million beautiful images. So what? As Matthew Berman incisively argues in "Human Creativity Still Matters," this abundance offers zero guarantee of meaning. While AI can produce visuals at scale, emotional impact remains a uniquely human currency, not an algorithmic output.

The abundance paradox reveals a critical flaw in equating volume with value. When production costs plummet, audiences drown in material, forcing creators into an even fiercer battle for attention. Berman highlights this: just because you can create "a million beautiful images doesn't mean that a million images are going to be felt emotionally by other people."

This necessitates a crucial test: human judgment. Work must meet people, and their reactions reveal its resonance. Berman emphasizes, "The only way to know whether a piece of art is going to be meaningful to humanity is to put it out there and see how people react." You can drop Mario into Elden Ring, but that’s recombination, not invention. Genuine innovation, the "next Mario" or "next Elden Ring," still demands human creativity to forge compelling stories and engaging experiences.

Matthew Berman, in his incisive “Human Creativity Still Matters” video, draws a crucial distinction between recombining existing elements and pioneering new ones. Generative AI excels at the former, brilliantly illustrating this with the hypothetical of Mario in Elden Ring. Yes, models can mash up beloved characters and worlds, rendering them with stunning fidelity. This is a remix, not a revolution.

Models learn from patterns in existing material; consequently, familiar styles and combinations are their native language. Crafting a new character, an unprecedented world, or a game paradigm demands more than sophisticated interpolation. It requires a human mind to conceive the "next Mario" or the "next Elden Ring"—concepts entirely outside the training data’s learned latent space.

Human vision remains paramount for innovation. While AI can certainly contribute, generating countless iterations or assets, a human still shapes which possibilities matter and why. You, the creator, provide the intent and the emotional compass, transforming algorithmic output into something that resonates. Without that human filter, even a million images offer zero guarantee of meaning.

When execution costs plummet, the creative advantage shifts dramatically. Matthew Berman rightly argues that if AI can churn out a million images for pennies, the true value isn't in making more, but in deciding what deserves to be made. This isn't about the quantity of assets, but the quality of the choices.

Taste becomes the new bottleneck, a working process of selection and refinement. It means choosing outputs that resonate, rejecting those that don't, and meticulously editing against a clear emotional or artistic intention. This isn't a passive appreciation; it’s an active, iterative loop of discernment that sculpts raw data into meaningful expression. For deeper insights into this dynamic, see Generative Artificial Intelligence, Human Creativity, and Art - PNAS / PubMed.

Curation builds audience trust. A deluge of generically polished assets feels like noise, but deliberate choices—a specific color palette, a unique character detail, an unexpected narrative twist—create a recognizable point of view. This distinct voice is what separates a pile of AI-generated content from something genuinely compelling, reflecting Berman’s assertion that "human creativity is more important than ever." You still need Human Creativity Still Matters, Human to decide what truly matters.

A compelling game transcends its asset library. True engagement stems from meticulous pacing, innovative mechanics, and balanced gameplay, all interwoven with a cohesive narrative. These are not merely visual triumphs; they are structural achievements, demanding human foresight and intuition to craft a truly fun and meaningful experience.

The real creative loop begins after launch. Developers release their vision, then observe how players engage—not just with surface-level metrics, but with the game’s emotional core and mechanical depth. This feedback, as Matthew Berman argues, is crucial: “The only way to know whether a piece of art is going to be meaningful to humanity is to put it out there and see how people react.”

For creators, this means treating AI as a powerful lever for ideation and iteration, not an autonomous author. Use AI to generate countless prototypes, test variations, and explore novel concepts rapidly. Ultimately, however, human accountability for the experience’s purpose and feel remains paramount, ensuring that innovation serves genuine player enjoyment.

Why does human creativity still matter when AI can generate art?

AI can produce and remix assets quickly, but people still decide what feels meaningful, original, and emotionally resonant.

Can AI create an entirely new game concept?

AI can help generate ideas and prototypes, but a compelling game still needs human direction, judgment, and choices about what makes play enjoyable.

What does “Mario in Elden Ring” illustrate?

It illustrates AI’s strength at recombining familiar elements—and why remixing them is not the same as inventing the next enduring idea.

How can creators use AI without losing their creative voice?

Use AI to speed up execution, then apply a clear point of view to select, shape, and test the work with real audiences.