It’s not just one thing — it’s another thing

Editor's take

A prevalent sentence structure, "It's not just X, it's Y," has emerged as a highly reliable indicator of AI-generated text, moving beyond a mere clue to near-certainty.

This linguistic tic, likely a byproduct of how large language models (LLMs) like GPT-3.5 and GPT-4 process and synthesize information, highlights the persistent challenge of achieving genuine stylistic naturalness in AI writing. The ubiquity of this pattern risks eroding trust in digital content, impacting everything from academic integrity to the perceived authenticity of news reporting and marketing materials.

Future developments will likely focus on LLMs' ability to diversify sentence construction and avoid repetitive phrasing. The success of models in this regard will be a key metric for their progress toward more human-like communication, and continued reliance on such predictable patterns could necessitate more sophisticated AI detection tools or prompt engineering strategies from users.