LingVo.club
Level
AI predicts personality from ordinary language — Level B2 — a black and white photo of the word mental health

AI predicts personality from ordinary languageCEFR B2

17 Feb 2026

Adapted from Jared Wadley-U. Michigan, Futurity CC BY 4.0

Photo by Marcel Strauß, Unsplash

Level B2 – Upper-intermediate
5 min
252 words

Researchers tested widely available generative AI models such as ChatGPT, Claude and LLaMa to see whether they can act as general judges of personality by reading ordinary language. The team used short daily video diaries and longer recordings from more than 160 people collected in both real-life and laboratory settings. Older text-analysis approaches did not perform nearly as well as these newer AI systems.

Results showed that AI personality scores closely matched how participants rated themselves and often matched self-ratings better than ratings from friends or family. The AI ratings also predicted meaningful outcomes: daily emotions, stress levels, social behaviour, and whether a person had been diagnosed with a mental health condition or had sought treatment. "We were taken aback by just how strong these associations were, given how different these two data sources are," says Aidan Wright, the study's first author.

Other scholars noted the implications. Chandra Sripada said the work supports the idea that language carries deep clues about psychological traits. Colin Vize described it as a new frontier in understanding human psychology, and Whitney Ringwald pointed out how personality appears in everyday experiences and passing thoughts.

  • Study limits: it relied on self-ratings and did not test direct comparisons with friends' or family judgments.
  • Researchers do not yet know whether AI and humans use the same signals or whether AI could outperform self-reports for major life outcomes like relationships, education, health or career success.
  • The findings appear in Nature Human Behavior; source: University of Michigan.

Difficult words

  • generativeable to create new content or data
  • personalityset of consistent psychological traits
  • diarypersonal record of daily experiences or events
    video diaries
  • self-ratingperson's own assessment of their traits
    self-ratings
  • predictsay something will happen based on evidence
    predicted
  • associationconnection or relationship between two things
    associations

Tip: hover, focus or tap highlighted words in the article to see quick definitions while you read or listen.

Discussion questions

  • How could AI personality assessments change the way people seek or receive mental health support? Give reasons.
  • Do you think everyday language gives enough information to judge personality accurately? Why or why not?
  • What ethical concerns might arise from using AI to assess people's personality from their language?

Related articles

Why many people stop exercising — Level B2
20 Jan 2026

Why many people stop exercising

Researchers found an “all-or-nothing” mindset makes people abandon exercise plans. Focus groups with adults showed strict rules, excuses, and puzzlement about keeping active. Experts suggest small thinking changes to help.