In practice, for many useful applications, many of the various obvious problems with AI agents (drift, hallucination, compounding errors) are more solvable than they are in theory
Clever prompting, tool use, constrained topics,
LLM judges & organizational process close some gaps— Ethan Mollick (@emollick) June 29, 2025
People are starting to sound more like AI chatbots in their everyday speech, according to a new study from the Max Planck Institute for Human Development. Researchers analyzed 280,000 academic videos across more than 20,000 channels and found that people are using words commonly found in AI-generated text, such as “meticulous,” “adept,” “delve,” and “realm,” much more frequently than before. These AI-favored words appeared over 50 percent more often than expected, showing that AI is having a real impact on the way people communicate.
The trend could make language less colorful and diverse, with less room for emotional nuance and regional differences.
All the technical language around AI obscures the fact that there are two paths to being good with AI:
1) Deeply understanding LLMs
2) Deeply understanding how you give people instructions & information they can act on.LLMs aren’t people but they operate enough like it to work
— Ethan Mollick (@emollick) June 28, 2025
There are also concerns about how talking to AI without using polite language could make people less friendly in their everyday conversations. At the same time, the convenience of using AI to write papers or meet deadlines can make its writing style tempting to copy.
Our latest speech-to-speech model is faster, more accurate, and excels at function calling. Watch @promptshant and @bfioca build a realtime voice agent that can search the web and hand off tasks to reasoning models with full context. pic.twitter.com/tLrSrceLFX
— OpenAI Developers (@OpenAIDevs) June 27, 2025
AI reshaping human communication patterns
Technology has always influenced language, from the telegraph making messages shorter to texting bringing acronyms and emojis into spoken language. But the current AI influence feels different.
It’s not just about efficiency or new words, but a growing sameness in how we express our thoughts. The irony is palpable: AI was designed to emulate human conversation, yet now humans are beginning to emulate AI,” said lead researcher Dr. Julia Krause.
“As this trend continues, individuals might need to consciously preserve the uniqueness and color of their speech to avoid the monotony of AI-like language.”
As AI becomes a bigger part of our lives, being aware of our word choices could be important for keeping human communication rich and diverse. The study’s findings highlight the need for further research into how AI is shaping language and the potential consequences for society.
Senior Software Engineer with a passion for building practical, user-centric applications. He specializes in full-stack development with a strong focus on crafting elegant, performant interfaces and scalable backend solutions. With experience leading teams and delivering robust, end-to-end products, he thrives on solving complex problems through clean and efficient code.
