Fraud-Spotting Data Analytics and AI Could Close £6 Billion Black Hole In Government Finances

Support our mission to provide fearless stories about and outside the media system

Packed with exclusive investigations, analysis, and features

SUBSCRIBE TODAY

Keir Starmer’s government could recoup billions of pounds by using digital technology and AI to tackle rampant fraud across Whitehall, a new report by the National Audit Office suggests today.

The disclosure comes after the Government was forced to drop plans to abolish the winter fuel allowance and cuts to disabled people, leaving around a £6 billion hole in Rachel Reeves spending plans. The move has led to speculation that she will have to raise taxes in the autumn Budget.

According to the NAO Whitehall is losing between £55 billion and £81 billion a year in fraud and error and the report says by using digital technology could prevent at least £6 billion of the losses every year in the immediate future. This is virtually equivalent to the money the Government has lost by restoring the winter fuel allowance and keeping PIP for disabled people. Given most of the cuts were not due to take place until the end of next year, it gives Whitehall time to plan.

Sir Geoffrey Clifton-Brown, Conservative Chair of the Committee of Public Accounts said: “With fraud and error potentially costing the taxpayer up to £80 billion a year, it is vital that Government prioritises data analytics to make sure the right amount of money goes to the right recipient. 

The £350 Billion a Year Lost to Corruption, Tax Dodging and Crime Every Year in the UK

As ministers search for cuts, a new parliamentary report reveals the hundreds of billions of pounds that could be clawed back into the UK’s flagging economy

Olly Haynes

“Whilst tools used in Government are often designed to detect transactions where fraud and error has already occurred, it is the preventative use that is much more cost-effective.”

The estimate comes from the Government Digital Service and is taken from savings already achieved by HM Revenue and Customs and the Department for Work and Pensions by using data analysis

The report says: “GDS believes Government could save as much as £6 billion a year by using data analytics to help tackle fraud and waste. The use of data analytics to tackle fraud and error has the potential to save billions of pounds of taxpayer money. Counter-fraud experts, within and outside of Government, consistently told us that data analytics needed to be a key part of any plan to reduce fraud and error.”

The DWP used data analysis to scrutinise Universal Credit claims and is estimated to have saved £588 million from potential fraud and error.

HMRC is using a number of in-house data analytics platforms to help decide where it should prioritise tax investigations, and to support those investigations. The platforms which were first introduced in 2010, now bring together over 100 datasets to find potential anomalies between an individual’s declared income and their assets or lifestyle. Savings amount to £3-4 billion a year.

Other projects include a pilot scheme in the NHS known as Project Athena, a data science and machine learning project, which tackles fraud. The pilot detects anomalous data points that could be prioritised for investigation or fraud prevention intervention, such as staff working elsewhere while claiming to be sick, or fake invoices.

EXCLUSIVE

Reform’s Darren Grimes Accused of Lying After Police Deny Telling Him Not to Hold Council Surgeries for ‘Security’ Reasons

EXCLUSIVE: Local police have rejected the Reform UK deputy council leader’s claim that they advised him not to meet his constituents

Josiah Mortimer

The ministry of Justice data science team has developed a machine learning model using AI that flags possible fraud, corruption and bribery using prison staff misconduct records. 

However the report states that it is not mandatory for every government department to introduce data analysis and there is no cross government strategy to implement it.

Gareth Davies, head of the NAO, said: “Government has the opportunity to make significantly more use of data analytics to tackle fraud and error, with potentially billions in savings available. It should learn from those who are doing this well, in both the private and the public sectors, to maximise the savings it can achieve.”  

A Government spokesperson said: “As these findings show, we’re making use of technology to crack down on fraud and error and have an ambitious plan to go further – from replacing legacy IT systems, to improving data sharing and scaling up the use of AI and data analytics.

“Departments are already deploying sophisticated AI tools alongside advanced data analytics to detect and prevent fraudulent activity more quickly and accurately. The National Fraud Initiative remains a cornerstone of this effort, using comprehensive data matching to uncover hidden fraud risks across the public sector.

“Tackling fraud and error is a vital priority and a key part of a wider programme of reform that we estimate could unlock savings of up to £45 billion over time by overhauling digital services.”