People are starting to notice that everyone’s talking more like AI lately. According to Sam Altman, the CEO of OpenAI, this isn’t just about using AI chatbots to write or communicate. It’s also about how even real human speech is sounding more “machine-like” and “fake.” Altman shared his thoughts on X (formerly Twitter), pointing out that many factors are pushing our language toward a more uniform, AI-influenced style.

He explains that people are adopting language patterns similar to those used by large language models, or LLMs. This shift is driven by online communities that tend to move together in their expression. Social media platforms often reward repetitive and exaggerated language because it boosts engagement. Creators also sometimes follow formulas that make their content more appealing, which further encourages this trend. Plus, actual bots are involved, making the overall language environment even more artificial. Altman believes these trends are making social media speech sound increasingly fake.

Science Backs Up the Change in Our Words

Research from UCLA and the University of Copenhagen supports the idea that AI influences how we choose words. The study looked at over 100,000 Reddit posts and comments made between April 2023 and January 2024. They found that certain words and phrases, like “delve,” “showcase,” and “underscores,” became more common after ChatGPT gained popularity. Many of these words weren’t used much before 2023, but their usage shot up once people became familiar with ChatGPT. Interestingly, this wasn’t just happening among AI users. The increase occurred even in casual conversations where people weren’t copying and pasting from chatbots. This shows that people are influenced by the language they see others use, spreading new words secondhand.

Another study points out that AI chatbots are also changing how we speak in real life. Human conversations often involve matching each other’s speech patterns, including tone, speed, and pronunciation. For example, if a New Yorker talks to a Texan, they tend to start sounding more like each other over time. As we interact more with AI voices that have their own accents and styles, we might start to imitate those patterns naturally. This could lead to a more homogenized way of speaking, where regional differences fade away.

The Broader Impact of AI on Language and Society

Altman’s concerns are valid: AI is reshaping our language, and that has big effects. One major issue is the potential loss of linguistic diversity. As AI favors clear, polished language, people might start adopting similar ways of speaking and writing, reducing the richness of dialects, slang, and cultural expressions. This could make global communication more straightforward but also more sterile and uniform. The unique quirks that give language personality and regional identity might disappear, replaced by a standard “GPT English.”

Language isn’t just a mirror of how we think; it also influences our thoughts. If AI-influenced language becomes the norm, it could shift how we process ideas and express ourselves. We might become more focused on clarity and efficiency, but lose the nuance, creativity, and storytelling that make human communication special. Conversations could feel less personal and more formulaic, shaped by the patterns favored by AI.

There are also social risks. If most AI voices favor a particular accent or style, it could reinforce biases and lead to subtle discrimination. People with regional accents might be viewed as less competent or less trustworthy. In education, students already use “GPT English” in their assignments, and teachers are noticing a decline in authentic voices. In workplaces, polished, machine-like language might become the standard, potentially marginalizing those who speak more casually or locally.

However, there’s a flip side. The influence of AI could also spark a desire for more genuine, diverse ways of speaking. If AI’s homogenized language becomes too dominant, people might push back by valuing more human, expressive communication. It’s a rapidly evolving situation, and we’ll see how language continues to change as AI becomes more integrated into our daily lives.

Inspired by

• https://www.computerworld.com/article/4055841/is-ai-changing-our-language.html

Sources

• gmpg.org
• scientificamerican.com
• arxiv.org
• machinesociety.ai
• superintelligentpodcast.com