AI Chatbots Influencing Human Speech Patterns: New Research Reveals Surprising Trends

How chatbots are making us speak like management consultants

A report from Florida State University revealed words such as "intricate" and "notable", which chatbots tend to use more than humans, have become more common in conversational podcasts since the release of ChatGPT in 2022. While the words have become increasingly common in written works such as scientific papers over the last two years, this has been attributed to authors using AI tools to write. However, the fact that they are also appearing in podcast conversations suggests AI is influencing how we speak, with more people echoing the sort of jargon often favoured by management consultants. The researchers have released their findings after analysing transcripts of popular science and technology podcasts, dividing them into two groups: those recorded between 2019 and 2021 and those since 2023. ChatGPT - the first widely used large language model - was introduced in November 2022. They found that several AI buzzwords were routinely used more often after the release of ChatGPT. The biggest rise was in "surpass", followed by "boast", "meticulous" and "strategically". "Given the short time frame and the number of words affected, this may indicate the onset of a remarkable shift in language use," the researchers said in a paper due to be presented at the Conference on AI, Ethics, and Society in October. Experts believe the AI systems may frequently turn to these words because of the way they are developed and later trained. Systems such as ChatGPT are built using a huge amount of text scraped from the web, a large portion of which includes corporate blog posts, annual reports and presentations, which tend to lean heavily on corporate jargon. They are then tweaked by a technique known as reinforcement learning from human feedback, in which human reviewers tell a model which version of an answer they prefer. Possible explanation for shift Some have suggested that, despite buzzwords rarely being used in spoken language, human reviewers respond well to them because they appear professional. The researchers said their study suggested that "human word choices have, moderately and selectively, converged with those of large language models". They said their sample covered 1,326 episodes of podcasts such as the BBC's Curious Cases science show and the Lex Fridman podcast, a popular technology show. Guests and hosts appearing on the podcasts were likely to be enthusiastic AI users than the average person, the researchers say, meaning the chatbot buzzwords are less likely to have seeped into other forms of speech, such as "rural sermons". A separate study of 360,445 YouTube videos and 771,591 podcasts last year also found an uptick in words frequently generated by ChatGPT, such as "delve", "comprehend", "boast", "swift", and "meticulous".

Researchers find evidence suggesting ChatGPT influences how we speak

Within five days of ChatGPT's release in 2022, the artificial intelligence chatbot gained more than a million users. 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. New research from Florida State University's Department of Modern Languages and Linguistics, Department of Computer Science, and Department of Mathematics is revealing notable evidence that AI buzzwords overused by chat-based large language models aren't just suggested on screens anymore -- they're frequently showing up in how people actually speak. Their work, "Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English," is the first peer-reviewed research to analyze whether the conversational human language system is changing in the wake of the widespread adoption of chat-based large language models like ChatGPT. The work is published on the arXiv preprint server. It was accepted into the eighth Conference on AI, Ethics, and Society, hosted by the Association for the Advancement of AI and Association for Computing Machinery, which will be held in October. The study will be published in EIES Proceedings as part of the conference. "This research focuses on a central issue in the discourse surrounding AI and language: are these language changes happening because we're using a tool and repeating what it suggested or is language changing because AI is influencing the human language system?" said assistant professor of computational linguistics and principal investigator Tom Juzek. "By analyzing lexical trends before and after ChatGPT was released in 2022, we found a convergence between human word choices and LLM-associated patterns with AI buzzwords." While rapid increases in the use of certain words -- like omicron -- do occur, these increases are typically due to real-world events. Recent large-scale upticks in the use of words like "delve" and "intricate" in certain fields, especially education and academic writing, are attributed to the widespread introduction of LLMs with a chat function, like ChatGPT, that overuses those buzzwords. "The changes we are seeing in spoken language are pretty remarkable, especially when compared to historical trends," Juzek said. "What stands out is the breadth of change: so many words are showing notable increases over a relatively short period. Given that these are all words typically overused by AI, it seems plausible to conjecture a link." Words including "surpass," "boast," "meticulous," "strategically," and "garner" have also seen considerable increases in usage since the release of ChatGPT. While these words are often used in a formal or academic tone, which makes them less common in unscripted, spoken language, researchers found that nearly three-quarters of these target words showed increased usage with some more than doubling in frequency. The team, including Bryce Anderson, a rising senior majoring in computer science, and Riley Galpin, a rising junior double-majoring in computer science and pure mathematics, analyzed 22.1 million words from unscripted and spontaneous spoken language, including conversational podcasts on science and technology. Post-ChatGPT, researchers noted a measurable increase in the use of AI-associated buzzwords but not their synonymous counterparts. For example, the AI buzzword "underscore" saw a considerable increase in usage while its synonym "accentuate" did not. FSU researchers said this shows that we're not just using AI; the language that AI uses is subtly seeping into how people talk. "Language is the most powerful medium of communication that humanity has and understanding how AI will affect this medium is fundamentally important and timely," said Anderson, the paper's lead author. "My motivation to pursue this research stems from seeing AI push the limits of what's possible in major industries and realizing that this influence isn't just limited to tool usage -- it can condition societal aspects, including how we use language." This research builds upon the group's previous research into the impact AI has on the domain of written scientific English, which revealed large-scale structural changes AI has had on scientific writing. Galpin presented the research on written language at the 38th International Florida AI Research Society Conference in May, and Anderson will join Juzek in presenting the team's new research on spoken language at the AIES Conference in October. "Our research highlights many important ethical questions," Galpin said. "With the ability of LLMs to influence human language comes larger questions about how model biases and misalignment, or differences in behavior in LLMs, may begin to influence human behaviors." This research was conducted as part of FSU's Undergraduate Research Opportunity Program, which provides high-achieving undergraduates an opportunity to explore their research interests alongside a faculty member. "I look forward to future research in this area," Juzek said. "While this is a critical development in understanding AI's role in language change, a big open question for future work is whether AI is amplifying conventional patterns of language change or directly driving them."

ChatGPT could be permanently changing how we speak

Just a few weeks ago, a paper from MIT stirred widespread debate on how using AI tools like ChatGPT was affecting our cognitive and critical thinking. The findings, though surprising, were not entirely unexpected. All the way back in 2023, Jeff Hancock, Professor of Communication at Stanford, revealed how AI chatter has already changed how we perceive language and their emotional impact. "What we found is that even if you don't use those AI-generated responses, they influence how you think," he said, citing the example of smart reply templates in emails and messaging apps. Two years later, it seems that the signature language patterns of large language models such as ChatGPT are becoming a part of natural human conversations, too. Recommended Videos The latest findings come courtesy of research from the Florida State University and it notes how "AI buzzwords overused by chat-based large language models aren't just suggested on screens anymore -- they're frequently showing up in how people actually speak." The team focused on a total of 20 words, and even I am surprised that at least two of them now appear regularly in my own work. What is happening? The team studied what they refer to as lexical trends before and after the release of ChatGPT, and discovered a theme of union between words chosen by people in their day-to-day conversations and those frequently used by AI chatbots. These words, such as "intricate" or "delve," have come to be known as AI buzzwords and are commonly seen in content generated by AI chatbots. The team assessed over 22 million words in spontaneous and non-scripted conversations such as podcasts covering technology and science-related topics. The researchers studied some famous podcasts such as BBC Curious Cases, Lex Fridman, and the Big Technology podcast, a list that totals 17 well-known audio shows. A noteworthy trend that was observed as part of the research was that AI buzzwords appear as such in human conversations, and their synonyms, which strengthens the belief that AI chatbots are influencing our speaking behaviour. But what's the bigger picture here? If you've been following the AI discourse, or just interact with AI chatbots such as ChatGPT or Gemini, you will notice that they use a few words quite frequently. There's a discernible pattern that is easy to spot, such as being more verbose than a casual human conversation, or even using too many em dashes in text. "What stands out is the breadth of change: so many words are showing notable increases over a relatively short period. Given that these are all words typically overused by AI, it seems plausible to conjecture a link," notes Tom Juzek, a professor of computational linguistics at the institution who led the latest research. The researchers found that words that are otherwise usually seen in academic work are popping up in human conversations, with some of them appearing with nearly double the usual frequency. The research poses the question of whether AI chatbots are having a "seep-in effect." More importantly, if AI buzzwords are already appearing in real human conversations, the researchers wonder how the inherent biases of an AI chatbot are going to affect human behavior in the long run. The research paper is the first to detail a link between the conversational human language system and how it is affected by the widespread usage of AI tools such as ChatGPT. The bigger picture Now, before you go thinking "whoever speaks these words must be talking too much with an AI," there are a few aspects you must keep in mind about this research. The team makes it clear that it's a case of "observed overlap in affected vocabulary," and doesn't speak of a definitive AI-induced language change. The paper, however, warns of a" novel and potentially profound shift in the trajectory of language evolution" if the preference for AI buzzwords keeps appearing in human communication. In the long run, the bigger risk is whether the biases of an AI chatbot also start appearing subconsciously in human conversations, even if they don't reflect the original thought process of the speaker. There is certainly precedent for that, as multiple cases have emerged where AI conversations have sent a person on a really dangerous path, leading to health emergencies and serious self-harm. The overarching takeaway of the research is that AI is not only morphing the language, but also essentially how we study language itself. Titled "Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English," the peer-reviewed paper will be published in the AIES Proceedings journal.

On-Screen and Now IRL: FSU Researchers Find Evidence Suggesting ChatGPT Influences How We Speak | Newswise

Newswise -- Within five days of ChatGPT's release in 2022, the artificial intelligence chatbot gained more than a million users. 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. New research from Florida State University's Department of Modern Languages and Linguistics, Department of Computer Science, and Department of Mathematics is revealing notable evidence that AI buzzwords overused by chat-based large language models aren't just suggested on screens anymore -- they're frequently showing up in how people actually speak. Their work, "Model Misalignment and Language Change: Traces of AI-Associated Language in Unscripted Spoken English," is the first peer-reviewed research to analyze whether the conversational human language system is changing in the wake of the widespread adoption of chat-based large language models like ChatGPT. It was accepted into the eighth Conference on AI, Ethics, and Society, hosted by the Association for the Advancement of AI and Association for Computing Machinery, which will be held in October. The study will be published in EIES Proceedings as part of the conference. "This research focuses on a central issue in the discourse surrounding AI and language: are these language changes happening because we're using a tool and repeating what it suggested or is language changing because AI is influencing the human language system?" said assistant professor of computational linguistics and principal investigator Tom Juzek. "By analyzing lexical trends before and after ChatGPT was released in 2022, we found a convergence between human word choices and LLM-associated patterns with AI buzzwords." While rapid increases in the use of certain words -- like Omicron -- do occur, these increases are typically due to real-world events. Recent large-scale upticks in the use of words like "delve" and "intricate" in certain fields, especially education and academic writing, are attributed to the widespread introduction of LLMs with a chat function, like ChatGPT, that overuses those buzzwords. "The changes we are seeing in spoken language are pretty remarkable, especially when compared to historical trends," Juzek said. "What stands out is the breadth of change: so many words are showing notable increases over a relatively short period. Given that these are all words typically overused by AI, it seems plausible to conjecture a link." Words including "surpass," "boast," "meticulous," "strategically," and "garner" have also seen considerable increases in usage since the release of ChatGPT. While these words are often used in a formal or academic tone, which makes them less common in unscripted, spoken language, researchers found that nearly three-quarters of these target words showed increased usage with some more than doubling in frequency. The team, including Bryce Anderson, a rising senior majoring in computer science, and Riley Galpin, a rising junior double-majoring in computer science and pure mathematics, analyzed 22.1 million words from unscripted and spontaneous spoken language including conversational podcasts on science and technology. Post-ChatGPT, researchers noted a measurable increase in the use of AI-associated buzzwords but not their synonymous counterparts. For example, the AI buzzword "underscore" saw considerable increase in usage while its synonym "accentuate" did not. FSU researchers said this shows that we're not just using AI; the language that AI uses is subtly seeping into how people talk. "Language is the most powerful medium of communication that humanity has and understanding how AI will affect this medium is fundamentally important and timely," said Anderson, the paper's lead author. "My motivation to pursue this research stems from seeing AI push the limits of what's possible in major industries and realizing that this influence isn't just limited to tool usage -- it can condition societal aspects, including how we use language." This research builds upon the group's previous research into the impact AI has on the domain of written scientific English, which revealed large-scale structural changes AI has had on scientific writing. Galpin presented the research on written language at the 38th International Florida AI Research Society Conference in May, and Anderson will join Juzek in presenting the team's new research on spoken language at the AIES Conference in October. "Our research highlights many important ethical questions," Galpin said. "With the ability of LLMs to influence human language comes larger questions about how model biases and misalignment, or differences in behavior in LLMs, may begin to influence human behaviors." This research was conducted as part of FSU's Undergraduate Research Opportunity Program, which provides high-achieving undergraduates an opportunity to explore their research interests alongside a faculty member. "I look forward to future research in this area," Juzek said. "While this is a critical development in understanding AI's role in language change, a big open question for future work is whether AI is amplifying conventional patterns of language change or directly driving them." To learn more about research conducted in the Department of Modern Languages and Linguistics, visit mll.fsu.edu.