[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-researchers-give-nine-ai-models-personality-tests-get-real-answers":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},7330,"researchers-give-nine-ai-models-personality-tests-get-real-answers","Researchers Give Nine AI Models Personality Tests, Get Real Answers","A new study tests nine AI models with psychometric instruments in English and Chinese, finding consistent, model-specific personality signatures.","Researchers just gave nine large language models a battery of personality tests, and the machines answered with distinct, repeatable personalities.\n\nResearchers built a cross-linguistic psychometric profiling framework and gave seven standard psychological instruments to nine LLMs, administering each test five times per model in both English and Chinese. When a model's answer could not be resolved after a retry procedure, it was logged as not applicable rather than discarded, and those refusals turned out to follow their own patterns rather than showing up at random. Every model shared an alignment-shaped baseline: high prosocial and self-regulatory scores, low dominance, disengagement and harmful-intent endorsement. Layered on top of that shared baseline, each model carried its own distinct profile, and repeated runs reproduced it closely enough that the researchers could identify which model produced a given set of answers.\n\nThat reproducibility matters more than the personality-test framing suggests. As LLMs increasingly draft emails, screen resumes and staff customer support chats, a repeatable way to measure a model's behavioral tendencies gives buyers and auditors something to check against, the same way a spec sheet lets you compare processors. It also means a model's answers shift with the language you address it in and which lab built it, a detail that matters for anyone assuming a chatbot behaves the same way across markets.\n\nNone of this means the models have inner lives. It means their alignment training leaves fingerprints consistent enough to show up on a psych test, which says more about the training process than about the thing taking the test.","[\"ai\",\"llm-research\",\"psychometrics\",\"alignment\"]","2026-09-23T04:00:00.000Z","2026-09-23T07:53:17.189Z","2026-09-23T07:53:22.186Z","published",null,[],"ai",[24,26,27,28],"llm-research","psychometrics","alignment",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.22934",0,{"sections":35},[36,39,43,48,53,58,62,67,72,77,82,87,92,97],{"name":37,"slug":24,"count":38,"latest_published_at":18},"AI",4281,{"name":40,"slug":41,"count":42,"latest_published_at":18},"Security","security",709,{"name":44,"slug":45,"count":46,"latest_published_at":47},"Policy","policy",369,"2026-09-23T02:13:52.000Z",{"name":49,"slug":50,"count":51,"latest_published_at":52},"Deals","deals",202,"2026-09-22T23:00:04.000Z",{"name":54,"slug":55,"count":56,"latest_published_at":57},"Hardware","hardware",168,"2026-09-22T23:56:03.000Z",{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",133,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",110,"2026-09-22T20:00:00.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":71},"Software","software",80,"2026-09-22T23:32:52.000Z",{"name":73,"slug":74,"count":75,"latest_published_at":76},"Dev Tools","dev-tools",79,"2026-09-22T22:21:13.000Z",{"name":78,"slug":79,"count":80,"latest_published_at":81},"Startups","startups",65,"2026-09-22T22:06:48.000Z",{"name":83,"slug":84,"count":85,"latest_published_at":86},"Gaming","gaming",45,"2026-09-22T15:35:06.000Z",{"name":88,"slug":89,"count":90,"latest_published_at":91},"General","general",43,"2026-09-21T23:48:56.000Z",{"name":93,"slug":94,"count":95,"latest_published_at":96},"Reviews","reviews",27,"2026-09-22T13:00:00.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]