AI Can Write. But It Can’t Teach Taste.
AI is evolving fast. But much of what it produces reflects the sensibility of an entry-level contributor: surface-level, broadly legible, and risk-averse. It can reword, remix, and scale content, but it can't yet discern what makes something exceptional, resonant, or true. That discernment comes from craft. And craft is inherited. Which raises the deeper risk: if experienced professionals exit the system instead of adapting to it, who teaches the next generation what good even looks like?
The Taste Transfer Gap
- Taste Requires Time: Taste isn't just preference. It's the result of deep exposure, critical feedback, and high standards over time. It's what separates average from excellent in editorial, strategy, and storytelling.
- AI Can Mimic, Not Mentor: Generative AI outputs reflect the average of what it ingests. It can scale what's already known, but it can't originate or teach judgment. Without human mentors to refine it, AI stays in the median.
- Generational Drift is Accelerating: Younger professionals are entering the workforce using AI as a baseline. But if their references are AI-trained rather than human-led, taste starts to flatten. The floor becomes the ceiling.
- The Knowledge Loop Breaks: When experienced writers, editors, and strategists opt out instead of leaning in, there's no feedback loop. AI iterates in a vacuum. Juniors train on the mediocre. Craft decays.
What It Looks Like in Practice
When Speed Outpaces Judgment
Across large enterprises, content teams are increasingly using AI to generate articles, reports, and marketing collateral. The speed is real, but so are the risks. Without experienced reviewers in the loop, teams are shipping outputs with hallucinated claims, fabricated data, or misaligned tone. These are not edge cases. They are the new baseline when judgment is removed from the process.
In strategy functions, teams using AI to accelerate slide development and research synthesis are facing a different failure mode: mediocrity. The slides are polished but lifeless. The narratives do not land. AI can format a story, but it cannot identify tension, decode subtext, or push an idea past the obvious. Without senior strategists shaping how prompts are constructed and how ideas are evaluated, the work may look impressive at first glance but lacks depth on inspection.
“I've reviewed hundreds of AI-written articles this year. The syntax is clean, but the soul is missing.”
What Junior Talent Actually Needs
Juniors don't need faster tools. They need exposure to judgment. A clear example of good. And access to people who can explain why one version of a message, story, or structure works better than another. Without this, they become prompt operators instead of communicators. They know how to get a fluent paragraph from ChatGPT, but not how to recognize if it’s insightful, on-brand, or strategically sound.
In previous generations, this came through apprenticeship. Listening to how senior editors rework a lead. Watching a strategist rewrite a slide to surface a core tension. Those invisible reps compound into skill. AI can’t provide that feedback loop unless someone builds it in.
Inside the Org: A Framework for Closing the Gap
- Craft Mapping: Identify where human taste still drives performance (e.g., messaging, editorial, brand tone). These are leverage points, not cost centers.
- AI Mentorship Programs: Pair experienced professionals with juniors not to resist AI, but to shape how it's used. Make taste transfer a deliberate function.
- Prompt Standards: Develop internal libraries of high-quality prompts and annotated outputs. Treat it like an editorial process, not a search engine.
- Quality Feedback Loops: Don’t just ship AI content. Review, refine, and document why certain versions are better. Treat feedback as craft inheritance.
Last Word
AI will not kill taste. But it will dilute it unless the people who know what good looks like stay in the loop. The future isn’t man versus machine. It’s mastery taught through machines. But only if the masters show up.
Need help building your organization’s taste infrastructure?
Let’s make AI a multiplier, not a ceiling. Contact →