[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-nhs-trained-a-national-ehr-ai-model-then-paused-its-own-access":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},5442,"nhs-trained-a-national-ehr-ai-model-then-paused-its-own-access","NHS Trained a National EHR AI Model, Then Paused Its Own Access","Researchers built a 243-million-parameter AI model on 61 million NHS patient records, but NHS England paused data access before results could be shared.","A UK research team trained a 243-million-parameter AI model on the medical records of roughly 61 million people - and then had its results locked away before anyone could see if the thing works.\n\nForesight-England is a generative model built from scratch inside NHS England's Secure Data Environment, reading de-identified patient timelines that combine primary and secondary care, death registrations, and COVID-19 data. It predicts a patient's next medical event autoregressively, working zero-shot across roughly 40,000 clinical codes drawn from ICD-10, OPCS-4, and SNOMED CT. The team trained on 54.9 million records from November 2018 to December 2022, held out 6.1 million for testing, and designed an evaluation for 30-day COVID-19 hospitalization and death risk - including checks against 2023 data the model never saw, benchmarked against plain logistic regression and XGBoost. NHS England paused the project's data access before those evaluations could run, so the paper describes only the architecture and method, not how well it performs.\n\nThat's the real story here. Building a foundation model at true national scale, rather than the single-hospital datasets most EHR models train on, is a genuine engineering feat. But a model nobody can evaluate is not evidence of anything - it's a blueprint with the results page torn out, and a reminder that governance can stall AI research even inside a government-run, purpose-built sandbox.\n\nCall it a national health AI experiment that got built, then got shelved before its own authors could grade it.","[\"ai\",\"healthcare data\",\"nhs\",\"foundation models\"]","2026-08-18T04:00:00.000Z","2026-08-18T20:23:54.761Z","2026-08-18T20:24:06.665Z","published",null,[],"ai",[24,26,27,28],"healthcare data","nhs","foundation models",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2608.16273",0,{"sections":35},[36,40,44,49,54,59,64,69,74,78,83,88,93,98],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",3293,"2026-08-20T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":39},"Security","security",435,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",210,"2026-08-19T09:32:27.000Z",{"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":58},"Hardware","hardware",140,"2026-08-19T18:25:42.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Consumer Tech","consumer-tech",95,"2026-08-18T16:05:00.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Science","science",90,"2026-08-19T18:41:02.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Software","software",73,"2026-08-18T07:51:50.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":18},"Dev Tools","dev-tools",69,{"name":79,"slug":80,"count":81,"latest_published_at":82},"Startups","startups",47,"2026-08-19T19:13:46.000Z",{"name":84,"slug":85,"count":86,"latest_published_at":87},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":89,"slug":90,"count":91,"latest_published_at":92},"General","general",33,"2026-08-18T22:18:13.000Z",{"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"]