Artificial intelligence (AI) is no longer just a tool for writing tasks, but also affects how we speak and interact with others. Researchers have documented shifts in the way people communicate as a result of ChatGPT, a language model that generates human-like responses.
The study analyzed over 280,000 YouTube videos from academic channels and found that speakers used words like “meticulous,” “delve,” and “realm” up to 51% more frequently than in the three years prior. These words align with those favored by ChatGPT.
Researchers believe that AI influence is not just limited to vocabulary but also affects tone, leading to longer, more structured speech and muted emotional expression. The study’s lead author, Hiromu Yakura, says, “Delve” has become an academic shibboleth, a neon sign in the middle of every conversation flashing ChatGPT was here.
However, this trend is not without concern. Researchers suspect that AI influence is starting to show up in the form of longer, more structured speech and muted emotional expression, which can lead to a loss of trust and authenticity in communication.
The study also found that AI systems actively flatten other dialects in ways that can demean their speakers, perpetuating inaccuracies about communities and what “correct” English is. This raises questions about preserving linguistic diversity and protecting the imperfections that build trust.
Ultimately, the future of AI voice depends on whether we’ll be conscious participants in its evolution. While there’s early backlash against homogenization, researchers believe that technology may evolve to better mirror human diversity rather than flatten it. The question remains: will we actively choose to preserve space for verbal quirks and emotional messiness that make communication recognizably human?
Source: https://www.theverge.com/openai/686748/chatgpt-linguistic-impact-common-word-usage