A Harvard student warns that AI writing is eroding what it means to think on the page

The New York Times dropped a critic and Hachette pulled a novel after both were found to involve AI writing. Readers are increasingly able to spot the pattern: polished prose, rehearsed emotion, no real surprise.

Categorized in: AI News Writers
Published on: Apr 12, 2026
A Harvard student warns that AI writing is eroding what it means to think on the page

AI-Generated Writing Has a Recognizable Voice-And Readers Are Starting to Notice

The New York Times has cut ties with a regularly contributing critic after the publication discovered the writer used AI in the process. A horror novel published by Hachette, one of the Big Five publishers, was pulled from shelves after readers and reviewers realized it had been written with AI assistance. These aren't isolated incidents-they signal a larger problem: AI-generated writing is becoming easier to spot, and harder to defend.

AI writes with a particular tone that repeats across platforms. It favors elaborate metaphors, rhythmic sentences, and emotional beats that feel rehearsed rather than earned. Read enough of it and the pattern becomes obvious. The writing lacks the friction that comes from a human mind working through difficulty on the page.

Consider the opening lines from the Hachette novel: "The ache is low and rhythmic, a second heartbeat in my ribs, steady and insistent, the kind of pain you get used to until it becomes part of you." The prose is polished. Every image lands. Nothing surprises. That's the problem.

Real writing-the kind that matters-arrives at its meaning through struggle. A writer wakes up with an idea that won't leave her alone. She rewrites the same paragraph seven times because something is still wrong. She discovers what she thinks only by writing it. AI cannot do this. It generates text based on patterns it has learned, not from lived experience or genuine discovery.

For writers, the stakes are practical. Tools like ChatGPT are now standard in many newsrooms and publishing houses. Editors face pressure to use them. Freelancers face pressure to compete with writers who do. The question becomes: How do you maintain your voice when the industry is pushing toward efficiency over authenticity?

One answer is to recognize what AI cannot replicate. It cannot write about what it means to fail. It cannot capture the specific texture of memory or the weight of a decision made at 3 a.m. It cannot surprise itself. These are the things human writers have always done best.

The risk is that constant exposure to AI-generated content will change how readers expect writing to sound. If enough emails, ads, and articles are written by AI, readers may begin to prefer that artificial polish. They may stop noticing when writing lacks ambiguity, complexity, or the marks of actual thought.

For writers who want to remain relevant and distinctive, the path forward is clear: lean into what makes your work human. Write the difficult thing. Pursue the idea that doesn't have an obvious answer. Let your voice stay rough where it needs to be rough. Understanding how AI tools work can help you recognize where they fail-and where your own work must go instead.


Get Daily AI News

Your membership also unlocks:

700+ AI Courses
700+ Certifications
Personalized AI Learning Plan
6500+ AI Tools (no Ads)
Daily AI News by job industry (no Ads)