[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-ai-pipeline-turns-ordinary-driving-footage-into-crash-tests":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},6905,"ai-pipeline-turns-ordinary-driving-footage-into-crash-tests","AI Pipeline Turns Ordinary Driving Footage Into Crash Tests","Researchers used computer vision, an LLM, and AR to insert staged crashes into real driving footage, testing self-driving systems without real-world risk.","A new automated pipeline takes ordinary, safe driving footage and digitally inserts a pedestrian crossing, a car cutting in, or a sudden stop, then hands the result to a self-driving system as a test.\n\nResearchers combined computer vision, a large language model, and augmented reality to build the system. It first tracks vehicles and pedestrians in a scene and calculates distance, speed, direction, and time-to-collision to judge how risky the moment already is. For scenes rated safe, an LLM invents a plausible collision-causing object or behavior, such as a jaywalker or a tailgating vehicle, and AR compositing drops it into the original footage. Tested on the nuScenes self-driving dataset, the system correctly classified scene safety 97.52% of the time and produced scenarios including pedestrian crossings, rear overtaking vehicles, and sudden-stop events.\n\nThis matters because the hardest part of testing autonomous vehicles is not building the car, it is finding enough near misses to prove the software will not fail on a rare, dangerous edge case. Real crash data is scarce, and collecting it on purpose is illegal and unethical, so simulation is the only realistic option. This approach turns an existing library of ordinary commute footage into a much larger set of near-crash tests, instead of relying entirely on synthetic scenes built from scratch.\n\nSimulated danger is still not the real thing, and no benchmark yet says whether a system that passes these AR-generated tests holds up on an actual street.","[\"autonomous-vehicles\",\"llm\",\"augmented-reality\",\"safety-testing\"]","2026-09-18T04:00:00.000Z","2026-09-18T21:58:10.333Z","2026-09-18T21:58:22.223Z","published",null,[],"ai",[26,27,28,29],"autonomous-vehicles","llm","augmented-reality","safety-testing",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.20318",0,{"sections":36},[37,40,44,49,54,58,62,67,71,76,81,86,91,96],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",4082,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",661,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",339,"2026-09-17T12:00:00.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":18},"Hardware","hardware",155,{"name":59,"slug":60,"count":61,"latest_published_at":18},"Science","science",125,{"name":63,"slug":64,"count":65,"latest_published_at":66},"Consumer Tech","consumer-tech",99,"2026-09-09T17:27:33.000Z",{"name":68,"slug":69,"count":70,"latest_published_at":18},"Dev Tools","dev-tools",78,{"name":72,"slug":73,"count":74,"latest_published_at":75},"Software","software",75,"2026-09-10T20:41:21.000Z",{"name":77,"slug":78,"count":79,"latest_published_at":80},"Startups","startups",55,"2026-09-09T23:14:29.000Z",{"name":82,"slug":83,"count":84,"latest_published_at":85},"Gaming","gaming",43,"2026-09-10T12:18:06.000Z",{"name":87,"slug":88,"count":89,"latest_published_at":90},"General","general",41,"2026-09-08T01:57:23.000Z",{"name":92,"slug":93,"count":94,"latest_published_at":95},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]