Human language is changing—and this time, the shift isn’t coming from pop culture, politics, or generational slang. It’s coming from AI. A new study from the Max Planck Institute finds that people are increasingly adopting AI language patterns, subtly absorbing the vocabulary and tone of tools like ChatGPT without realizing it.

This transformation is happening across YouTube, Reddit, corporate messaging, and even Parliament floors—raising a deeper question: Are humans shaping AI language, or is AI already shaping ours?

TL;DR

A Max Planck Institute study shows that everyday language is evolving to mirror AI tools like ChatGPT. People on platforms such as YouTube and Reddit are increasingly using rare or formal words (“underscore,” “meticulous,” etc.), while moderators struggle to distinguish AI-generated posts from human ones. This shift is also showing up in political speeches and corporate notices, revealing a feedback loop where humans copy AI, and AI learns from humans—making both sound increasingly alike.

What Did the Max Planck Study Discover About AI Language Patterns?

Researchers analyzed the way people communicate online post-ChatGPT and found a noticeable rise in words rarely used in casual conversation before 2023.

The New Vocabulary: Chatbot Words Go Mainstream

Terms like:

• “underscore”
• “comprehend”
• “meticulous”
• “bolster”

have spiked in usage on YouTube. These are words that appear frequently in AI responses but were never common in everyday speech.

The implication is straightforward: Users are parroting the tone of AI tools, often unconsciously.

Why Are Reddit Moderators Seeing “AI-Like” Human Posts?

Online moderators, especially on large human-interest subreddits, have been sounding the alarm for months. They say posts now read like neatly packaged, polished paragraphs straight out of ChatGPT.

Humans Are Writing Like Bots

According to moderators:

• Posts once messy and emotional are suddenly “perfectly structured.”
• Tone and cadence mimic AI’s signature calm, orderly style.
• Distinguishing genuine human writing now requires “gut instinct.”

What makes this tricky is that many of these posts are human-written—people are simply internalizing patterns from the content they consume.

This isn’t just a platform issue; it’s a linguistic evolution happening in real time.

How Are AI and Humans Creating a Language Feedback Loop?

The study points to a powerful loop shaping modern communication:

AI Learns From Us: Then We Learn From AI

One Reddit moderator explained it cleanly:
“AI learns from people, and people copy what they see online.”

This cycle works like this:

1. Humans write online.
2. AI models train on that writing.
3. AI adopts a particular tone—polite, orderly, and slightly formal.
4. Humans imitate that tone because it now dominates online spaces.
5. AI trains on that, reinforcing the cycle.

The result?
Human and machine writing styles are slowly converging.

AI Language Patterns Are Starting to Show Up in Public Speech

It’s not just online speech that’s shifting. Public officials are picking up AI-sounding phrasing, even when they don’t intend to.

UK Lawmakers Using an Unusual, AI-Like Phrase

Essayist Sam Kriss found that British lawmakers have begun saying “I rise to speak,” a phrase ubiquitous in American political scripts but uncommon in UK parliamentary tradition.

The phrase was used 26 times in a single day, a noticeable spike.

It’s unclear whether this came directly from AI-generated documents, online cross-pollination, or simply cultural drift. But the tone feels unmistakably “LLM-like”: formulaic, formal, and oddly detached.

Corporate Communications Are Starting to Sound Like They Were Written by Bots

If you’ve walked past a closed store lately and wondered whether the sign was written by a person or ChatGPT… you’re not alone.

Starbucks Closure Notices Spark Suspicion

When Starbucks shut some stores earlier this year, the door signs used heavily sentimental, flowery prose, phrasing many now associate with generative AI.

Examples included:

• elaborate gratitude statements
• soft, emotional tone
• “We cherish this community,” style wording

It’s still unclear whether a bot wrote them or a human mimicking a bot did. And that uncertainty is exactly the point.

As corporate communication becomes more polished and sanitized, it increasingly resembles the default tone of AI—making it harder to tell the difference.

Why Does This Shift in AI Language Patterns Matter?

Language shapes how we think, build relationships, and understand the world. When a dominant tone—especially one created by machines—begins to influence our default communication style, several questions emerge:

• Are we losing linguistic diversity as writing becomes more standardized?
• Will younger users adopt AI-influenced vocabulary as their natural voice?
• Could institutions gradually adopt AI-like formality in speech and policy documents?
• How much of our language is still ours?

For linguists, this moment is unprecedented: never before has a non-human system contributed so directly to daily vocabulary and tone.