[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-models-flunk-an-audiovisual-art-reasoning-test":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},2760,"ai-models-flunk-an-audiovisual-art-reasoning-test","AI Models Flunk an Audiovisual Art Reasoning Test","A new benchmark finds the best multimodal AI scores just 48% on questions about creative intent in film, visual art, and games — humans average 87%.","The best AI in the room still can't explain why a director cuts to silence.\n\nResearchers introduced MuseBench, a 4,016-question benchmark testing whether multimodal large language models understand the *why* behind artistic choices — not just what appears on screen or canvas. The questions span cinema, static visual art, stage performance, and game design, and were distilled from more than 10,000 video essays pairing professional critical commentary with visual examples. A four-phase pipeline screened out shortcuts and added adversarial distractors before expert validators signed off. Twenty-eight current models were evaluated in zero-shot conditions.\n\nThe results are blunt: the top-performing model hit 48.29% accuracy. Human experts averaged 87.18%. That nearly 39-point gap matters because it targets something most benchmarks skip — reasoning about creative intent, like why grief is conveyed through silence and a lingering close-up rather than a tearful monologue. Existing benchmarks mostly test perceptual recognition, which models have largely cracked; MuseBench tests interpretation, which they haven't.\n\nThe timing is awkward for labs racing to pitch multimodal models as creative collaborators. If a model can't reason about why a filmmaker chose claustrophobic framing to amplify fear, its usefulness for genuine artistic analysis is limited — no matter how fluently it describes what it sees. MuseBench gives the field a harder target to aim at, and right now, nothing is close.","[\"ai\",\"multimodal\",\"benchmarks\",\"creative-ai\"]","2026-06-30T04:00:00.000Z","2026-06-30T12:14:17.965Z","2026-06-30T12:14:20.870Z","published",null,[],"ai",[24,26,27,28],"multimodal","benchmarks","creative-ai",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.30026",0,{"sections":35},[36,40,45,50,55,60,65,70,75,80,85,89,94,99],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":76,"slug":77,"count":78,"latest_published_at":79},"Dev Tools","dev-tools",59,"2026-07-07T04:00:00.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":86,"slug":87,"count":83,"latest_published_at":88},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]