[{"data":1,"prerenderedAt":138},["ShallowReactive",2],{"uc-org-department-for-work-and-pensions":3},{"organization":4,"includeUnpublished":14,"evidence":15},{"slug":5,"name":6,"country":7,"region":8,"industry":9,"wikidata":10,"records":11,"useCases":12,"indexable":13},"department-for-work-and-pensions","Department for Work and Pensions","GB","europe","government","Q1191394",3,4,true,false,[16,62,106],{"title":17,"useCases":18,"organization":21,"vendors":22,"summary":26,"stage":27,"year":28,"channels":29,"languages":31,"metrics":33,"outcomeDisclosed":13,"sources":44,"verification":52,"grade":55,"id":56,"useCaseTitles":57},"Department for Work and Pensions: machine learning risk model for Universal Credit advances",[19,20],"benefit-fraud-and-error-detection","application-and-identity-fraud-detection",{"name":6,"anonymized":14,"country":7,"region":8,"industry":9,"wikidata":10},[23],{"name":24,"role":25},"In house (Integrated Risk and Intelligence Service)","in-house","DWP scores requests for Universal Credit advances in real time with a supervised machine learning classifier and refers the highest risk requests to a caseworker before payment. The caseworker is not told the referral came from the model, a random control group is referred alongside, and every decision to decline is made by a person and can be appealed. DWP's published effectiveness assessment for April 2025 to March 2026 finds the model 2.5 times more effective than random selection with a median payment delay of one day for approved referrals. It also finds that non UK nationals and several age bands were referred more often without a matching increase in confirmed fraud, and that referrals of couples were less often confirmed than those of single claimants; a retrained model was being tested.","scaled",2026,[30],"api",[32],"en",[34],{"kpi":35,"value":36,"unit":37,"qualifier":38,"period":39,"baseline":40,"claimant":41,"quote":42,"sourceUrl":43},"detection-rate-improvement",2.5,"multiplier","exact","1 April 2025 to 31 March 2026","Randomised control group sample of advances","organization","The performance information for 2025 to 2026 demonstrates the model is 2.5 times more effective at identifying fraud risk than a randomised control group sample.","https:\u002F\u002Fwww.gov.uk\u002Fgovernment\u002Fpublications\u002Feffectiveness-assessment-including-statistical-analysis-of-the-universal-credit-advances-machine-learning-model-1-april-2025-to-31-march-2026\u002Feffectiveness-assessment-of-universal-credit-advances-model",[45,48],{"url":43,"title":46,"publisher":6,"date":47},"Effectiveness Assessment of Universal Credit Advances Model","2026-07-09",{"url":49,"title":50,"publisher":51},"https:\u002F\u002Fwww.gov.uk\u002Falgorithmic-transparency-records\u002Fdwp-universal-credit-advances-model","DWP: Universal Credit Advances Model (algorithmic transparency record)","GOV.UK",{"level":53,"checkedAt":54},"source-verified","2026-09-26","B","dwp-universal-credit-advances-fraud-model",[58,60],{"slug":19,"title":59},"AI for benefit fraud and error detection in social security",{"slug":20,"title":61},"AI for application and identity fraud detection",{"title":63,"useCases":64,"organization":66,"vendors":67,"summary":74,"stage":75,"year":76,"channels":77,"languages":79,"metrics":80,"outcomeDisclosed":13,"sources":96,"verification":100,"grade":55,"id":102,"useCaseTitles":103},"DWP Child Maintenance Service: predicting payment breakdown on compliant cases",[65],"propensity-to-pay-scoring",{"name":6,"anonymized":14,"country":7,"region":8,"industry":9,"wikidata":10},[68,71],{"name":69,"role":70},"Tata Consultancy Services","integrator",{"name":72,"role":73},"Amazon Web Services (SageMaker)","platform","The Child Maintenance Service (CMS) in the Department for Work and Pensions (DWP) uses a machine learning model to predict which paying parents on compliant Collect and Pay cases are likely to miss a payment. A model trained each month on case, payment, liability and contact data (an XGBoost classifier with SHAP explanations) scores around 100,000 cases and shows caseworkers a high, medium or low risk with the reasons and suggested actions, so they can step in before a payment breaks down. Caseworkers decide what to do; the score is one input. The tool went to a wider group of caseworkers in March 2025 and DWP reports validation accuracy of 0.86 and precision of 0.79.","production",2025,[78],"internal-tools",[32],[81,89],{"kpi":82,"value":83,"unit":84,"qualifier":85,"period":86,"claimant":41,"quote":87,"sourceUrl":88},"interactions-handled",100000,"count","approximately","cases scored each month","The tool is used at the start of each month to predict the risk of non-compliance, around 100,000 cases are processed by the model and presented to the CMS2012 System.","https:\u002F\u002Fwww.gov.uk\u002Falgorithmic-transparency-records\u002Fdwp-child-maintenance-service-cms-predictive-analytics-compliance",{"kpi":90,"value":91,"unit":92,"qualifier":38,"period":93,"baseline":94,"claimant":41,"quote":95,"sourceUrl":88},"accuracy",86,"percent","validation data from May 2025","The model is accepted only when precision on validation data is at least 0.62; the record reports precision of 0.79.","Accuracy: 0.86 Precision: 0.79",[97],{"url":88,"title":98,"publisher":51,"date":99},"DWP: Child Maintenance Service (CMS) Predictive Analytics Compliance (algorithmic transparency record)","2025-10-30",{"level":53,"checkedAt":101},"2026-10-13","dwp-child-maintenance-payment-breakdown-prediction",[104],{"slug":65,"title":105},"AI propensity to pay scoring for arrears and debt collection",{"title":107,"useCases":108,"organization":110,"vendors":111,"summary":114,"stage":27,"year":76,"channels":115,"languages":117,"metrics":118,"outcomeDisclosed":13,"sources":128,"verification":133,"grade":55,"id":134,"useCaseTitles":135},"DWP: Conversational Platform on the benefits telephone lines",[109],"benefits-eligibility-and-application-assistant",{"name":6,"anonymized":14,"country":7,"region":8,"industry":9,"wikidata":10},[112],{"name":113,"role":73},"Omilia Natural Language Solutions","Callers to the benefit telephone lines of the Department for Work and Pensions (DWP) are asked what they are calling about. Speech recognition and a natural language model then signpost them to GOV.UK, help them log in to their online account, answer some questions in the IVR (such as the next payment amount and date) or route them to the right adviser. The platform also runs identity and verification questions against the DWP trust hub by API. It makes no decisions; callers can always ask for a human.",[116],"voice",[32],[119,124],{"kpi":82,"value":120,"unit":84,"qualifier":85,"period":121,"claimant":41,"quote":122,"sourceUrl":123},1000000,"calls per month","At present approximately 1 million calls per month pass through the Conversational Platform.","https:\u002F\u002Fwww.gov.uk\u002Falgorithmic-transparency-records\u002Fdwp-conversational-platform",{"kpi":90,"value":125,"unit":92,"qualifier":38,"period":126,"claimant":41,"quote":127,"sourceUrl":123},97,"speech recognition success","97% speech recognition success",[129],{"url":123,"title":130,"publisher":131,"date":132},"DWP: Conversational Platform","GOV.UK (Algorithmic Transparency Recording Standard)","2025-12-16",{"level":53,"checkedAt":54},"dwp-conversational-platform",[136],{"slug":109,"title":137},"AI assistant for benefits eligibility questions and applications",1791904366145]