Universal Credit advances model saves an estimated £4.4 million

AI system

System: Universal Credit advances machine learning model
Model: Not assessed
Release: Not assessed
Application developer: Department for Work & Pensions (DWP)
Role: Not assessed
Configuration: Not assessed
Inputs and outputs: Not assessed
Autonomy: Not assessed
Workflow: A model trained on historical fraud and error data in advance claims. Referrals above a threshold are reviewed by caseworkers alongside control-group referrals with scores removed.
Human oversight: Not assessed
Relation to AI: The NAO describes a model trained on historical fraud and error data that scores advance requests and refers those above a threshold. Every payment decision is made by a caseworker, so the model's contribution is prioritization. Developer and deployer are the same public body (DWP). The NAO, an independent auditor, wrote the report.
AI contribution: The NAO describes a model trained on historical fraud and error data that scores advance requests and refers those above a threshold. Every payment decision is made by a caseworker, so the model's contribution is prioritization. Developer and deployer are the same public body (DWP). The NAO, an independent auditor, wrote the report.

Outcome identifiers: Not assessed
Benefit: saving an estimated £4.4 million from 2022–23 to 2024–25 by DWP's estimate
Category: Reduced Risk
Status: Not assessed
Beneficiary groups: Organizations. Society.
Beneficiaries: Not assessed
Reach: Not assessed
Distribution and access: Not assessed
Duration: Not assessed
Expected?: Yes
Costs and harms: Not assessed

Studies: Not assessed
Study design: Not assessed
Measurement: Not assessed
Setting and sample: Not assessed
Measures: saving an estimated £4.4 million from 2022–23 to 2024–25 by DWP's estimate
Comparison: a randomised control group sample of advance requests
Results: Reference: fraud-risk identification rate of the random control sample (not quantified). Result: around three times more effective at identifying fraud risk.
Additional value from AI: Not assessed
Limitations: Not assessed
Denominators: Not assessed
Materials for replication: Not assessed

Kind of value: advances that would otherwise have been overpaid, saved from 2022-23 to 2024-25
Amount: Estimate: £4.4 million
Currency and prices: GBP, identified from the source context. Nominal source amount; price year not separately stated.
Period and unit: U.K. fiscal years 2022–23 through 2024–25. Aggregate Universal Credit advances model estimate.
Realization: Retrospective estimate.
Recipients and payers: Recipient: U.K. Department for Work and Pensions and public funds. Payer: DWP benefit expenditure that would otherwise have been overpaid.
Gross or net: Estimated gross overpayments avoided. DWP could not separate this model’s costs from wider analytics costs.
Included costs: Not assessed
Comparison: DWP’s estimate of payments that otherwise would have been made; method is not detailed in the cited passage.
Method: Not assessed
Return and payback: Not assessed

Participants: Developer: Department for Work & Pensions (DWP). Operator: Department for Work & Pensions (DWP).
Jurisdictions: Developer: United Kingdom. Operator: United Kingdom.
Ownership and control: Not assessed
Dependencies: Not assessed
Cross-border connections: Not assessed
Geopolitical implications: Not assessed

Conditions for replication: A human decision on every advance. Regular fairness assessments (age is the only protected characteristic in the model). Feedback of review outcomes into the model.
Replication attempts: 2 follow-up entries. Benefit recurred: 1. Outcome unresolved: 1.
Later findings: 9 July 2026. DWP's effectiveness assessment for April 2025 to March 2026 found the retrained model 2.5 times more effective than a random control group and concluded it should keep operating.
Availability: Not assessed
Institutional decision: Not assessed
Grounds: Not assessed
Action and result: Not assessed
Success criterion: Not assessed

Assessed benefits: Organizations. Society.
Depth: Not assessed
Reasons: Organizations: The saving is an aggregate departmental estimate over three years, not an attributable outcome. Society: No outcome for claimants or taxpayers beyond the aggregate estimate is documented. Referred claimants bear an added evidence burden.
Observed or intended: Not assessed
Rubric: Not assessed
Assessor and date: Not assessed
Alternative assessments: Not assessed
Fill: Independently studied

Kind: Not assessed
Related entries: Not assessed
Field: Benefit fraud and error. Universal Credit advances.
Purpose: Assess the fraud risk of Universal Credit advance requests and refer high-risk requests to a caseworker for review.
Dates: Date basis unassessed: 2025 (precision unassessed). Updated: 21 October 2026.
Stage: Not assessed
Version: SHA-256: 537199b2c3da…

Report Independent author

National Audit Office, 'Tackling benefit overpayments due to fraud and error', 22 October 2025, paragraphs 15-16 and 3.4-3.15
Report Deployer

Department for Work & Pensions (2026). Effectiveness Assessment of Universal Credit Advances Model, 1 April 2025 to 31 March 2026. GOV.UK, 9 July 2026.
News report Independent author

Jim Dunton, Civil Service World, 22 October 2025