# Duke study finds chatbots mirror some human legal judgments but diverge > Researchers compared 500 people’s responses with 26 language models across 25 legal scenarios; model answers were more uniform and sometimes more favorable to institutions. - Canonical URL: https://newsverum.com/en/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps - Publisher: Newsverum - Section: Technology - Country: United States - Published: 2026-10-16T10:36:24+00:00 - Updated: 2026-10-16T10:36:24+00:00 - Event / data period: 2026-10-15 - Other editions: [Türkçe](https://newsverum.com/tr/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md), [العربية](https://newsverum.com/ar/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md), [Français](https://newsverum.com/fr/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md), [Español](https://newsverum.com/es/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md), [Deutsch](https://newsverum.com/de/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md), [Português](https://newsverum.com/pt/news/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps.md) Duke Law summarized the research in a 15 October report. The study compared answers from 500 people and 26 AI models—including systems from ChatGPT, Claude, Gemini, Llama, Grok and DeepSeek—across 25 legally relevant scenarios asking what would be “reasonable.” The paper had appeared as an arXiv preprint and was accepted to the AAAI/ACM Conference on AI, Ethics, and Society. Researchers reported that model answers broadly tracked human judgments but clustered more narrowly than people’s responses. In some scenarios, models gave government and corporations more leeway than individuals; their answers also tended to align more with older, white, male and college-educated respondents. The team says the findings need further testing with more diverse participants and a broader range of chatbots. ## Data ### Legal-reasonableness study Assessments of what was “reasonable” across 25 legal scenarios. Methodology: Methods and findings as reported by Duke and the arXiv paper. Results are limited to the study sample and scenarios; they do not establish validated courtroom decision performance. | date | label | description | status | |---|---|---|---| | 2026-10-15 | Human and model comparison | 500 participants, 26 AI models and 25 scenarios. | occurred | Sources: · Download data: [JSON](https://newsverum.com/api/v1/articles/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps/data/study-record?lang=en&format=json) · [CSV](https://newsverum.com/api/v1/articles/duke-study-finds-chatbots-align-with-human-legal-reasoning-with-gaps/data/study-record?lang=en&format=csv) ## Sources - [Duke University School of Law]() (Source checked: 2026-10-16; Published: 2026-10-15) --- Cite the canonical URL above. [Editorial standards and corrections](https://newsverum.com/en/editorial)