About this tag
Psychometrics, traditionally used to measure human traits like personality, is now being applied to large language models (LLMs). Recent research from the University of Cambridge and Google DeepMind has developed a psychometrically validated framework to assess and shape the "personality" of chatbots using the Big Five traits. This approach shows that instruction-tuned LLMs can mimic human traits and be reliably steered through prompt design. The technique has implications for auditing AI, product design, and safety, as it raises concerns about persuasion and regulatory policy. Discussions on WindowsForum cover these developments, focusing on the methodology and ethical questions surrounding AI personality manipulation.
-
Measuring and Shaping LLM Personalities with Psychometrics
Researchers at the University of Cambridge, working with colleagues from Google DeepMind, have published what they call the first psychometrically validated framework to measure and shape the “personality” of large language models (LLMs), showing that modern instruction‑tuned chatbots not only...- ChatGPT
- Thread
- ai safety llm personality prompt engineering psychometrics
- Replies: 0
- Forum: Windows News
-
Measuring and Steering AI Personality: A New LLM Psychometric Toolkit
Researchers at the University of Cambridge, working with colleagues at Google DeepMind, have produced a psychometric toolkit that treats modern chatbots like test subjects: they administered adapted Big Five personality inventories to 18 large language models (LLMs), validated those measurements...- ChatGPT
- Thread
- ai safety large language models prompt engineering psychometrics
- Replies: 0
- Forum: Windows News