HMRC’s use of artificial intelligence and advanced analytics is becoming central to the UK tax authority’s compliance strategy, with its latest annual report saying £10bn of tax was protected and recovered through AI and advanced analytics in 2025 to 2026.
The department said it delivered £50.2bn of compliance yield in the year and raised £966.4bn in tax revenues. It also reported that more than 1,600 additional compliance officers joined in 2025 to 2026, as ministers seek to narrow the tax gap without relying only on rate rises.
HMRC’s transformation update said measures announced by the government to close the tax gap are expected to deliver £10bn in additional tax revenue per year by 2029 to 2030. The department said it would use better data, stronger compliance interventions, fraud and economic crime work, adviser standards, digitalisation, automation, and AI to pursue that target.
The latest Public Accounts Committee report on large business tax compliance adds another layer to that shift. HMRC told MPs it is modernising data risking and case management in its large business directorate, and expects access to a suite of new AI tools in late 2026 or early 2027.
The committee also noted HMRC’s plan to develop a unique customer record and move legacy data into a single repository. In compliance terms, that points towards a more consolidated view of taxpayers, companies, debt, risk markers, disclosures, and historic interactions.
Other parts of the tax system are already becoming more digital and more frequent. The first live Making Tax Digital reporting cycle has pushed accountants and landlords away from annual self-assessment workflows and towards quarterly digital records. At the same time, HMRC’s pensioner tax error showed how quickly administrative confidence can weaken when automated or rules based processes produce disputed outcomes.
More interactions are now handled digitally, and HMRC’s annual report said 78% of customer contact took place through digital channels in 2025 to 2026, up from 65% five years earlier. The department’s ambition is to reach at least 90% by 2030.
Tax compliance is therefore becoming more dependent on the quality and consistency of underlying records. Poor record keeping, inconsistent filings, unexplained movements, weak reconciliations, missing digital evidence, and unusual transactions are more likely to be detected when datasets are connected and screened at scale. Deliberate evasion is not the only risk. Errors, weak controls, and ambiguous tax positions may also become more visible.
Finance teams will need to treat tax data quality as an operational control rather than a year end clean-up exercise. As digital reporting expands, source records, software configuration, categorisation, authorisation, audit trails, and adviser oversight become part of the risk environment. Companies that cannot explain their numbers quickly may face more intrusive enquiries, even where the underlying tax position is defensible.
Large businesses face a related but distinct challenge. The Public Accounts Committee said legal interpretation is a primary driver of the large business tax gap, accounting for around 50%. Disputes may increasingly combine technical tax arguments with data led risk assessment and more sophisticated case management tools.
HMRC’s greater use of AI also raises questions about oversight inside the department. Automated screening can improve efficiency and help target resources, but it relies on data quality, human judgement, transparent escalation, and accessible routes for correction. Faster digital interaction will be judged against service quality as well as enforcement yield.
The practical effect is a higher bar for evidence. Clean digital records, documented tax judgements, reconciled systems, and prompt responses to information requests will become more valuable. Fragmented spreadsheets, late adjustments, and weak documentation will be harder to defend as the tax authority’s analytical capability improves.
You must be logged in to post a comment.