AI Slop Is So Now. But AI Quality Starts 2,000 Years Ago.
In the age of generative tools, distinguishing between hollow "AI slop" and meaningful content has become a defining challenge for creators. "AI slop" is easy to recognize: it is generic, statistically average, and says a lot while meaning very little. But evaluating and maintaining quality in your own AI-assisted work requires a deeper standard—one rooted in principles over two millennia old.
That's part of why I wanted to do this series. At Metamorphosis, we use AI too, in a lot of what we build and create, including The Playbook. But we're constantly trying to do that with ever-improving quality and scalability, not just more output. So if that's what bad AI looks like, what does good AI look like? What actually is quality, especially in the age of AI?
Where Quality Actually Comes From
The search for quality begins with Aristotle and his concept of poion—the category of what kind of thing something is. For Aristotle, a thing’s quality was inextricably linked to its purpose (telos). A knife is high-quality if it cuts well, because cutting is its reason for being. This idea is the origin of modern "fit for purpose" standards.
Centuries later, Robert Pirsig explored this concept in Zen and the Art of Motorcycle Maintenance. Pirsig argued that quality isn't easily captured by rigid logic alone; you feel it before you rationalize it. You recognize quality intuitively before you can fully explain why it works.
Why AI Broke the Traditional Definition
For years, quality creative work in marketing and advertising was defined as work that is "fit for purpose and elicits an emotional response that yields action."
That standard was built for a world of human-only creation, where the primary risks were blandness or self-indulgent irrelevance. Generative AI completely changes the risk landscape. Because AI defaults to the median of its training data, the primary risk now is producing homogenized, soul-less content at scale.
To ensure AI enhances rather than dilutes quality, work must meet four distinct criteria:
Specificity: Low-quality AI produces generic output meant for everyone, which resonates with no one. High-quality AI work becomes increasingly specific—tailored to a single voice, audience, or purpose so tightly that it could not apply to anyone else.
A Surviving Point of View: AI should act as a contributor, not a replacement. Human discernment, judgment, and lived experience must shape the work so that a clear human perspective remains evident in the final result.
Human Judgment Over Passive Approval: AI excels at synthesis, research, and overcoming blank-page paralysis, but it cannot know what truly matters to a specific audience. Quality drops the moment humans stop making creative calls and begin simply rubber-stamping AI output.
Multiplying Leverage, Not Replacing Thought: High-quality AI use compresses time and expands access without compressing critical thinking. It should multiply a person’s existing expertise rather than substitute for a lack of it.
Criterion One: Specificity
Criterion Two: Does a Point of View Survive?
Criterion Three: Judgment Versus Approval
Criterion Four: Multiply, Don't Replace
A Modern Definition of Quality
Updating the classic standard for the AI era yields a comprehensive definition:
Quality creative—whether AI-assisted or not—is work that is fit for purpose, elicits an emotional response that yields action, and expands a specific person's judgment rather than replacing their experience.
Ultimately, AI should not replace human judgment; it should operate in service of it. The benchmark for AI tools is not merely whether they make creation faster, but whether they make the final outcome more authentic, sharp, and aligned with the human behind it.
Comments