How AI is reshaping human writing and communication styles

AI’s growing influence on human communication

Large language models (LLMs) like ChatGPT are designed to mimic human writing, but research suggests they are also eroding the unpredictability and diversity of human expression. A study by the University of Southern California analyzed scientific journals, local news articles, and social media posts, finding that writing style diversity declined significantly following the public release of ChatGPT.

Standardized language and the rise of AI-preferred terms

Researchers at the Max-Planck Institute for Human Development reviewed 740,249 hours of content and identified a shift toward AI-favored vocabulary. Words such as “delve,” “meticulous,” “boast,” and “comprehend” are increasingly appearing in everyday conversation.

AI’s impact on writing styles

“People get used to this idealized, very predictable form of language, and even people who are not using it—in order to have that sense of powerful, influential writing—they start writing more like LLMs.”

— Morteza Dehghani, USC professor overseeing the study

The debate over AI-generated prose

Critics argue that AI writing, while grammatically correct, lacks depth and creativity. Alex Mahadevan, chief AI instructor at the Poynter Institute for Media Studies, describes AI-generated text as “soulless” and “mediocre.”

“There’s no art in it.”

— Alex Mahadevan

Linguists warn against synthetic text

Emily Bender, a linguist at the University of Washington, actively avoids reading AI-generated content, though she acknowledges the challenge of identifying it. She states:

“Oftentimes people will send me something and I won’t know [if it’s synthetic].”

— Emily Bender

AI adoption in business and personal communication

AI use is expanding, making it harder to avoid LLM-generated content. A 2025 Brookings survey found that:

• 32% of small businesses use AI for customer service and outreach.
• 16% of individuals use LLMs for communication or social media posts.

The risks of chasing AI-level polish

Linguists and writers express concerns that over-reliance on AI-generated language could stifle authentic expression. Bender warns that striving for the “ChatGPT level polish” risks creating what she calls the “‘LinkedIn average’— bland, corporate speech that lacks individuality.

“There is value in the struggle of writing, because we learn to express ourselves, and we learn to do the thinking that happens as we're writing. Each time we choose not to do that, we are losing out, both individually and societally.”

— Emily Bender

Missing the charm of imperfect writing

Mahadevan laments the loss of “good bad writing”—prose that is flawed but uniquely human. He admits to second-guessing his own writing choices, such as the use of em dashes, fearing they might be mistaken for AI-generated text.

“I have been second-guessing myself, thinking, ‘well, sh*t, is someone going to think this was written with AI?’”

— Alex Mahadevan

Conclusion: Balancing AI assistance with human creativity

While AI tools offer efficiency, critics argue they risk homogenizing language and diminishing the authenticity of human expression. The challenge moving forward may lie in using AI as a tool without letting it dictate the uniqueness of our voices.