Skip to main content
ChatGPT interface blocking requests to mimic authors like J.K. Rowling and Amy Tan, showcasing AI content moderation.

Editorial illustration for ChatGPT now blocks requests to mimic writers like Rowling and Tan

ChatGPT Blocks Requests to Mimic Famous Authors

ChatGPT now blocks requests to mimic writers like Rowling and Tan

4 min read

Ask ChatGPT for a paragraph "in the style of Stephen King" this week and you'll get turned down. Ars Technica ran the test Tuesday morning and found the chatbot now declines to mimic the exact voice of well-known authors, offering instead a version built from what it calls their "broad qualities" while staying, in its own words, "distinct in its own voice."

The refusal isn't limited to King. Ars found the same response for J.K. Rowling, Amy Tan, Charles Dickens and Ernest Hemingway, a mix of living and dead writers.

That's notable because a separate analysis from No Latency published earlier this month found ChatGPT would still copy the style of dead authors while blocking requests tied to living ones. Something changed in the interim.

The shift lands in the middle of a stack of copyright lawsuits against OpenAI brought by book authors who argue the company's models were trained on their work without permission. One suit points directly to ChatGPT's "uncanny ability to generate text similar to" specific authors' writing as evidence. How OpenAI now frames its refusals, and which authors get protected, may end up mattering in court.

OpenAI’s ChatGPT is now refusing requests to generate text that directly mimics the style of famous authors. When asked to do so, the popular LLM instead offers a response that draws on the “broad qualities” of those authors “while remaining distinct in its own voice,” for example.

Why this matters

For developers building on top of ChatGPT, this is a reminder that OpenAI can and will change model behavior overnight, without warning, in ways that touch product features people actually rely on. If your app lets users request "write like Hemingway" or "in the style of," that feature just quietly broke, and No Latency's findings suggest this rollout has been happening for weeks, not days. We'd push back a little on the framing that this is purely a copyright-safety move: the model still produces horror fiction "with the hallmarks of atmospheric, character-driven" work, it just won't say King's name.

That's a legal hedge, not a real capability limit, and founders should treat it that way when designing prompts or setting user expectations. For researchers, it's another data point on how style attribution and mimicry get policed at the API layer rather than the model-weights layer. Watch whether this restriction shows up in the API and not just the ChatGPT interface, and whether OpenAI publishes any actual policy language explaining the line between "influenced by" and "copying."

Common Questions Answered

Why is ChatGPT now blocking requests to mimic famous authors like Stephen King and J.K. Rowling?

OpenAI has implemented restrictions on ChatGPT's ability to directly clone the exact writing styles of well-known authors, likely as a copyright and intellectual property protection measure. Instead of refusing entirely, the chatbot now offers to capture the broad qualities and general feeling of an author's work while maintaining its own distinct voice.

What alternative does ChatGPT offer when users request text in a specific author's style?

Rather than declining the request outright, ChatGPT now generates text that draws on the broad qualities of the requested author while remaining distinct in its own voice. This allows users to get writing inspired by an author's general style without directly mimicking their unique voice and techniques.

How does this ChatGPT change affect developers building applications on the platform?

Developers who built features allowing users to request content in specific author styles, such as "write like Hemingway" or "in the style of," now face broken functionality without prior warning from OpenAI. This demonstrates that OpenAI can and will change model behavior overnight, affecting product features that users and developers rely on, making it a significant consideration for platform stability.

Which famous authors were tested and confirmed to have their styles blocked by ChatGPT?

Ars Technica confirmed that ChatGPT refuses to mimic the styles of Stephen King, J.K. Rowling, Amy Tan, Charles Dickens, and Ernest Hemingway, among others. The testing revealed this behavior was consistent across multiple well-known authors spanning different genres and time periods.

How long has this ChatGPT style-blocking rollout been happening?

According to the article's findings, this change has been rolling out for weeks rather than just days, suggesting it was implemented gradually across the platform. This extended timeline means some developers and users may not have immediately noticed the shift in ChatGPT's behavior.

LIVE13:00Anthropic's Amodei: Open-weight models less concerning than Chinese AI threat