I’ll believe that from German researchers when LLMs start shoehorning four nouns together instead of just delving.
A team from the Max Planck Institute for Human Development in Germany published a non-peer-reviewed preprint copy of research they say detects that words that ChatGPT uses preferentially have started to appear more frequently in human speech since the bot was unleashed on the world in 2022.
So-called “GPT words” include comprehend, boast, swift, meticulous, and the most popular, delve. After analyzing 360,445 YouTube academic talks and 771,591 podcast episodes, the team concluded words like delve, swift, meticulous, and inquiry were just a few examples of terms that began appearing in more podcasts and videos across various topics.
I’d like to nominate “authenticity” as the vapid word of the year.
Alternate title: Generation that
venerateddisregarded books reading for the first timeedit: omg for some reason I thought venerated meant dislike (it means the opposite!!). This is what happens when you’re given books at too young an age folks. You pull fancy words out of your ass cuz they vibe right. But you risk realizing that you probably inferred the wrong definition from wherever you learnt it.
I know this feeling all too well. I was told about context clues and handed books from a very young age. It was sometimes a decade or two before I found out my presumed definition of a word was wrong. More regularly, though, I tried to show off my newfound vocabulary and was quickly disabused of my error.
I think people’s vocabularies got worse as they stopped reading. While irritating, this is nothing compared to that.
I’ve heard as a possible explanation before that LLMs are mainly trained on long-form written texts, whereas more colloquial speech (spoken or in text chat form) typically uses simpler speech. So, where normal humans would put “Sure!” or “Yeah!”, the LLM will likely write “Certainly!”, because that’s more likely what’s written in a blog post.
