AI-Associated Language Changes: New Research Reveals Shifts in Human Speech Patterns

Strategic Insights -

As AI chatbots like ChatGPT continue to gain widespread adoption, new research from Florida State University's Department of Modern Languages and Linguistics, Department of Computer Science, and Department of Mathematics is shedding light on the impact of AI on human language. Within five days of its release in 2022, ChatGPT gained over a million users, and today, more than half of all adults under age 30 and nearly half of adults ages 30-49 say they've used it, according to Pew Research Center. The research, "Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English," analyzed 22.1 million words from unscripted spoken language and found notable evidence that AI buzzwords are seeping into how people actually speak.

Key Takeaways:

  • The study found that within five days of ChatGPT's release in 2022, it gained over a million users.
  • Over 50% of adults under age 30 and nearly 50% of adults ages 30-49 have used ChatGPT, according to Pew Research Center.
  • Researchers analyzed 22.1 million words from unscripted spoken language and found a measurable increase in the use of AI-associated buzzwords.
  • Words including "surpass," "boast," "meticulous," "strategically," and "garner" saw considerable increases in usage since the release of ChatGPT.
  • The team found that nearly three-quarters of the target words showed increased usage with some more than doubling in frequency.
  • Researchers noted a measurable increase in the use of AI-associated buzzwords but not their synonymous counterparts, suggesting that AI is influencing human language.
  • The study highlights important ethical questions about the influence of AI on human language and behaviors.

Statistics:

  • 22.1 million words from unscripted spoken language were analyzed in the study.
  • Over 50% of adults under age 30 and nearly 50% of adults ages 30-49 have used ChatGPT.
  • The AI buzzword "underscore" saw a considerable increase in usage, while its synonym "accentuate" did not.
  • Nearly three-quarters of the target words showed increased usage with some more than doubling in frequency.
  • The study found a measurable increase in the use of AI-associated buzzwords between pre- and post-ChatGPT analysis.

Sources:

  • Florida State University
  • Pew Research Center
  • Association for the Advancement of AI and Association for Computing Machinery (Conference on AI, Ethics, and Society)
  • EIES Proceedings
  • Undergraduate Research Opportunity Program
  • Department of Modern Languages and Linguistics, Department of Computer Science, and Department of Mathematics, Florida State University