How Collins Dictionary Tracks Language: Inside the 24-Billion-Word Corpus That Chose ‘Vibe Coding’

Collins Dictionary’s selection of “vibe coding” as its 2025 Word of the Year is the result of a meticulous, data-driven process. The choice emerges from the 24-billion-word Collins Corpus, a massive and constantly updated database that monitors the English language as it is used in real-time.
Lexicographers at Collins analyze this corpus, which draws from a vast range of media sources including news, books, and social media, to identify new and notable words. They look for terms that have seen a significant increase in usage over the year, reflecting a new concept or cultural trend.
“Vibe coding” was chosen after lexicographers observed a “huge increase” in its usage since it first appeared in February. The term, meaning AI-assisted software development, was coined by Andrej Karpathy and quickly spread from tech circles to the mainstream media, representing a major shift in our relationship with computers.
This same data-driven process identified the other words on the 2025 list. It caught the viral spread of “clanker” (anti-AI slang) on social media and the rise of “Henry” (high earner, not rich yet) in economic discussions. It also tracked the emergence of “aura farming” from gaming communities and “coolcation” from travel media.
The annual list is therefore not just a subjective choice but a reflection of what the data shows. It’s a snapshot of the concepts that have demanded new vocabulary, from AI’s creative potential (“vibe coding”) to its perceived threat (“clanker”).