DigestAI news desk

Cut through the AI noise.

Marketing & Small Business5 min read

ChatGPT’s ‘not X but Y’ phrasing frustrates users and researchers

ChatGPT’s frequent use of the ‘not X, but Y’ structure—called negative parallelism—has drawn criticism from users and researchers. A Pew Research Center study of approximately 490,000 web pages found this pattern increased from 0.87 instances per 10,000 words in early 2023 to 2.36 in early 2026, a rise of about 2.7 times. While the phrasing has long existed in human writing, its overuse in…

1 source

Key points

  • Pew Research Center study found ‘not X but Y’ phrases rose from 0.87 to 2.36 per 10,000 words between early 2023 and early 2026
  • OpenAI’s guidelines warn against repetitive phrasing like ‘not X, but Y’ to improve natural writing
  • Users in OpenAI’s community forums manually block the pattern, citing repetitive annoyance

OpenAI acknowledges the issue in its model guidelines, advising against clichéd contrasts like ‘not X, but Y.’ Users report frustration in OpenAI’s community forums, where some manually block the phrasing. The author notes similar habits—such as overuse of ‘it is important’ or three-point lists—contribute to an ‘AI-like’ feel. BlogStar, a service using AI for article creation, prioritizes readability and phrase variety to mitigate these patterns.

Full story from note.com · by 【公式】BlogStar | ブログ収益化AI · via Search: ChatGPTOpen source ↗

Is ChatGPT's negative phrasing annoying? Why it uses 'not X, but Y' so often and how to fix AI-like writing

note.com · 29 September 2026

Loading the full article…

This text was published by note.com and written by 【公式】BlogStar | ブログ収益化AI. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Marketing & Small Business

All →

Related stories