[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-physstream-lets-you-steer-ai-video-physics-as-it-renders":10,"sections":35},{"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":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},6432,"physstream-lets-you-steer-ai-video-physics-as-it-renders","PhysStream Lets You Steer AI Video Physics As It Renders","A new research model lets you nudge object motion while an AI-generated video is still rendering, using physics signals instead of pixel dragging.","An AI video generator that lets you nudge how objects move while the clip is still rendering, rather than locking in every physics detail before you hit generate.\n\nResearchers describe PhysStream, an autoregressive image-to-video model built for physics-grounded scenes. It tracks objects and their positions across previously generated frames, building a structured scene memory it references as it renders. Instead of dragging pixels around, users feed it sparse velocity-increment signals, actual physical quantities, so the model has to learn real dynamics rather than fake convincing motion. The team trained it in two stages, first tuning a bidirectional model on motion control, then adding the scene-memory component to a causal autoregressive version, and tested it on multi-object tabletop rigid-body scenes.\n\nMost controllable video tools force a tradeoff: either you commit to a full control script before generation starts, or you get pixel-level control that pushes objects to specific spots without any sense of momentum, collision, or weight. PhysStream's mid-generation, physics-based control is a meaningful step past both approaches, and it shows in the numbers: a 33% drop in motion distribution distance and 12% lower trajectory error versus the strongest baselines, with human evaluators favoring it in over 85% of head-to-head comparisons.\n\nStill, this is tabletop objects in synthetic benchmarks, not open-world video, so the real test is whether physics-grounded control holds up in messier scenes: cluttered rooms, cloth, water, crowds.","[\"ai video generation\",\"generative ai\",\"physics simulation\",\"research\"]","2026-09-16T04:00:00.000Z","2026-09-17T16:02:32.625Z","2026-09-17T16:02:44.531Z","published",null,[],"ai",[26,27,28,29],"ai video generation","generative ai","physics simulation","research",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.17521",0,{"sections":36},[37,41,46,51,56,60,64,69,74,78,83,88,93,98],{"name":38,"slug":24,"count":39,"latest_published_at":40},"AI",3852,"2026-09-17T08:27:09.000Z",{"name":42,"slug":43,"count":44,"latest_published_at":45},"Security","security",648,"2026-09-17T04:00:00.000Z",{"name":47,"slug":48,"count":49,"latest_published_at":50},"Policy","policy",338,"2026-09-11T04:00:00.000Z",{"name":52,"slug":53,"count":54,"latest_published_at":55},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":57,"slug":58,"count":59,"latest_published_at":45},"Hardware","hardware",154,{"name":61,"slug":62,"count":63,"latest_published_at":45},"Science","science",114,{"name":65,"slug":66,"count":67,"latest_published_at":68},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":45},"Dev Tools","dev-tools",73,{"name":79,"slug":80,"count":81,"latest_published_at":82},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":84,"slug":85,"count":86,"latest_published_at":87},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":92},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":94,"slug":95,"count":96,"latest_published_at":97},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":99,"slug":100,"count":101,"latest_published_at":102},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]