Thousands of Britons could face HMRC investigations as the revenue authority intensifies its crackdown on unpaid tax, using artificial intelligence and data analytics to identify tax dodgers more easily than ever. The evolving approach helped HMRC secure or protect around £10 billion during the last financial year, reports suggest.
AI and data analytics drive investigations
Connect, HMRC's data analysis system, cross-checks information from banks, social media platforms, online marketplaces and tax returns. This system led to 540,000 investigations during the 2024/25 year, according to law firm Pinsent Masons. Ian Robotham, partner at the firm, said: "The algorithms that it uses allow HMRC to spot anomalies that would otherwise go unnoticed by the human eye."
Increased willingness to prosecute
Ian Dickinson, tax director at UHY Hacker Young, added: "HMRC seems increasingly willing to prosecute individuals who evade tax." The authority's Fraud Investigation Service records 260 convictions secured against the most serious tax evaders during the 2025/26 year.
Whistleblower rewards scheme boosted
HMRC is also boosting its network of informants by offering a bolstered rewards scheme, with whistleblowers receiving between 15% and 30% of the extra tax collected when providing information that leads to over £1.5 million being retrieved. Individuals informing on their neighbours, colleagues and friends were handed a total of £1.4 million after a record number of tip-offs in the last financial year, according to government data.
Expectation of more whistleblower reports
Hinesh Shah of Pinsent Masons said: "We're expecting that the number and quality of whistleblower reports to HMRC will increase dramatically under the new reward systems - which is based on the US system." In the US, $123.5 million was given to tax whistleblowers as rewards in the most recent year for which data is available. "If HMRC starts issuing individual awards that go into the millions of pounds then that is going to lead to a radical change in behaviour," Shah added.
