HM Revenue & Customs has selected SAP to modernise its core tax systems, placing artificial intelligence at the heart of how the UK manages revenue, compliance, and taxpayer services. The move reflects a broader shift in public-sector digital strategy — from layering AI on top of ageing systems to redesigning infrastructure that can support automation natively.
The UK’s tax system is one of the most complex administrative environments in the world. Each year, HMRC manages more than £800 billion in tax revenue, supports over 45 separate tax regimes, and serves millions of individuals and businesses across the country. Behind this scale sits a sprawling technology landscape that has evolved over decades, often through incremental upgrades rather than fundamental redesign.
That approach is now changing.
By selecting SAP to overhaul its core revenue infrastructure, HMRC is signalling that the future of tax administration will be driven by cloud-native platforms, unified data, and AI-enabled decision-making — not retrofitted automation layered onto legacy systems.
Moving Beyond Incremental Modernisation
For years, many public-sector bodies approached digital transformation cautiously. AI tools were often introduced as add-ons, designed to automate isolated tasks without touching the underlying architecture.
While this reduced risk in the short term, it also limited impact.
“AI struggles to deliver value when it’s forced to operate on fragmented, ageing infrastructure,” one government technology advisor observed. “You can automate the surface, but the complexity underneath remains.”
HMRC’s partnership with SAP represents a different philosophy. Instead of working around legacy constraints, the tax authority is replacing the core platform itself to support machine learning, analytics, and automation as built-in capabilities.
This approach treats AI not as a feature, but as an outcome of modern infrastructure.
The Role of the Enterprise Tax Management Platform
At the centre of the transformation is HMRC’s Enterprise Tax Management Platform (ETMP) — the system responsible for processing, managing, and analysing tax data across the UK.
ETMP is not a single application, but a foundational platform that supports everything from tax calculations and compliance to reporting and enforcement. Tens of thousands of HMRC employees rely on it daily to carry out critical functions.
Over time, ETMP has become increasingly complex, reflecting changes in legislation, policy, and taxpayer behaviour. Maintaining this complexity across on-premise systems has made it difficult to introduce advanced analytics or automation at scale.
By migrating ETMP to a managed cloud environment through RISE with SAP, HMRC aims to simplify that landscape and create a foundation where AI capabilities can operate effectively.
Why Cloud Is a Prerequisite for AI in Tax Systems
Machine learning depends on consistent, high-quality data. Models trained on incomplete or siloed datasets produce unreliable outputs — an unacceptable risk in tax administration.
Traditional on-premise environments often struggle to provide the data unification AI requires. Different systems store data in different formats, with varying governance rules and update cycles.
SAP’s cloud-based approach addresses this by centralising data management and standardising how information is accessed and analysed.
As part of the deployment, HMRC will implement SAP Business Technology Platform (BTP) alongside embedded AI capabilities. These tools are designed to:
- Surface insights more quickly
- Automate routine processes
- Support advanced analytics and forecasting
- Enable real-time reporting
For HMRC staff, this means faster access to reliable data and greater confidence in analytical outputs.
AI as an Enabler, Not a Replacement
Despite the scale of the transformation, HMRC has been clear that AI is intended to support decision-making, not replace human judgment.
Tax administration involves nuance, discretion, and legal interpretation. AI systems can help identify patterns, flag anomalies, and prioritise cases — but final decisions remain with trained professionals.
This human-in-the-loop approach is especially important in public-sector contexts, where transparency, fairness, and accountability are non-negotiable.
By embedding AI into core systems rather than bolting it on, HMRC aims to ensure that automation operates within clearly defined boundaries.
Data Governance and the Importance of Sovereign Cloud
Deploying AI in tax administration raises immediate questions around data security, sovereignty, and compliance. Tax data is among the most sensitive information any government holds.
To address this, HMRC will host its modernised platform on SAP’s UK Sovereign Cloud, ensuring that data residency and governance requirements are met.
Sovereign cloud environments allow public bodies to use commercial cloud technology while maintaining control over:
- Where data is stored
- Who can access it
- How it is processed and audited
This balance is critical for public trust.
“Large-scale public systems must operate reliably at national scale while adapting to changing demands,” said Leila Romane, Managing Director UKI at SAP.
“By modernising one of the UK’s most important platforms and hosting it on a UK sovereign cloud, we are strengthening the resilience, security, and sustainability of critical national infrastructure.”
Building AI on a Trusted Foundation
For AI to be accepted in public services, trust must be built into the system architecture itself.
That includes:
- Clear data lineage and audit trails
- Explainable AI outputs
- Strong access controls
- Compliance with UK regulatory frameworks
SAP’s sovereign cloud model supports these requirements while allowing HMRC to benefit from continuous platform updates and security enhancements.
This approach reflects a growing consensus in government IT: AI adoption must be grounded in governance first, innovation second.
Improving the Taxpayer Experience Through AI
While much of the transformation focuses on internal systems, the long-term goal is to improve interactions between HMRC and taxpayers.
Complex tax processes often result in delays, confusion, and uncertainty for individuals and businesses. AI-enabled systems can help reduce friction by:
- Speeding up processing times
- Improving the accuracy of assessments
- Providing clearer, more timely information
- Supporting more consistent decision-making
SAP and HMRC will jointly define AI use cases that prioritise these outcomes, ensuring that automation aligns with service quality rather than efficiency alone.
Empowering HMRC Staff With Better Tools
The modernisation also changes the daily experience of HMRC employees.
Legacy interfaces often require staff to navigate multiple systems to complete a single task. This slows productivity and increases the risk of error.
The new SAP-based platform is designed to provide:
- Improved user interfaces
- Unified access to analytical data
- Faster reporting and case management
With better tools, staff can focus on higher-value work such as complex investigations, policy analysis, and taxpayer support.
For public-sector leaders, this highlights an often-overlooked benefit of AI modernisation: workforce enablement.
Lessons for Other Public-Sector Organisations
HMRC’s approach offers insights for other government bodies exploring AI adoption.
The project demonstrates that AI is as much an infrastructure challenge as a software one. Without addressing technical debt, data fragmentation, and governance constraints, AI initiatives struggle to scale.
Key takeaways include:
- Modernise core platforms before deploying AI
- Prioritise data quality and accessibility
- Secure sovereign or compliant cloud foundations
- Embed governance into system design
These principles apply well beyond tax administration, from healthcare to social services and beyond.
Avoiding the Pitfalls of Legacy Automation
One of the risks in public-sector automation is using AI to mask outdated processes rather than redesign them.
HMRC’s strategy avoids this by rethinking how tax systems operate end-to-end. Instead of automating inefficiencies, the goal is to simplify workflows before applying AI.
This reduces long-term maintenance costs and ensures that automation delivers sustainable value.
A Signal of Maturing Public-Sector AI Adoption
The HMRC-SAP partnership reflects a broader maturation in how governments approach AI.
Early experimentation has given way to more deliberate, infrastructure-led strategies. Rather than chasing innovation headlines, public bodies are focusing on reliability, resilience, and measurable outcomes.
This quieter phase of AI adoption may be less visible, but it is ultimately more impactful.
What Comes Next
As HMRC migrates its tax infrastructure to SAP’s cloud platform, the focus will shift from implementation to optimisation.
Over time, AI models can be refined, new use cases introduced, and processes continuously improved — all within a stable, governed environment.
For taxpayers, the benefits may appear gradually: faster responses, clearer communication, and fewer administrative hurdles.
For government leaders, the project sets a benchmark for how to modernise critical national systems responsibly.
Reframing AI as National Infrastructure
Perhaps the most important takeaway from HMRC’s transformation is how it reframes AI’s role.
In this context, AI is not a novelty or a standalone initiative. It is an extension of national infrastructure — something that must be secure, resilient, and designed for decades of use.
By partnering with SAP and prioritising cloud-native architecture, data governance, and sovereign deployment, HMRC is laying the groundwork for an AI-enabled tax system that can evolve alongside the UK’s economy.
In doing so, it offers a model for how public institutions can adopt AI with ambition — and with care.