Artificial intelligence is rapidly reshaping the tax landscape, and this transformation was front and centre at the ITR AI in Tax Forum in London in September 2025. The event brought together tax leaders, technology experts and innovators to explore how AI is redefining compliance, risk management and operational efficiency across the tax function. While adoption is still in its early stages, the message was clear: AI is moving fast, and businesses that start preparing today will lead tomorrow.
With growing regulatory complexity and reliance on vast data sets, integrating AI into tax functions is becoming essential. Practitioners at the event emphasised the importance of human oversight, rigorous quality checks and well-defined use cases for successful AI implementation.
Here are our key takeaways from the event and practical insights to help your organisation navigate the shift toward an AI-driven tax future.
Key takeaways
Early adoption
Early adoption of AI works best when humans are kept in the loop. Organisations implementing AI should develop a clear strategy, align tax AI initiatives with broader company goals for AI adoption and start small.
Best practices include:
- Start with narrow, focused tasks to gain quick wins
- Maintain expert checks by tax and finance specialists
- Define mandatory quality checks to ensure accuracy
- Measure outcomes such as time saved, error reduction and process efficiency
Workflows & production
When integrating AI into your workflows and production, start with the problem you want to solve and the outcome you want to achieve. Too often, organisations begin with the AI technology and work backwards. This leads to misaligned priorities and wasted effort.
Recommendations include:
- Break complex processes into smaller, specialised AI agents that each handle a focused task. This approach promotes clarity, performance improvement and measurable results.
- Avoid building large, cross-functional AI systems initially to prevent “pilot fatigue”.
- Focus on outcome-driven AI workflows for smoother implementation
Data quality, literacy & process
A major theme throughout the event was the growing recognition that AI in tax is only as good as the data behind it. Firms are investing heavily in automation and analytics, yet many still struggle with inconsistent data sources, manual handoffs, and fragmented systems. Poor data quality undermines confidence in insights, slows decision-making, and increases compliance risk. The message was clear: before implementing AI, organisations must first establish a foundation of clean, structured, and accessible data. This isn’t just about technology, it’s about governance, ownership, and a commitment to accuracy at every stage of the tax lifecycle.
Equally important was the emphasis on repeatable processes and data literacy. Tax teams must design processes that are standardised, automated, and traceable to ensure consistency across jurisdictions and reporting periods. However, process alone isn’t enough, tax teams need to understand the data they work with. Strengthening data literacy enables professionals to question anomalies, interpret analytics and confidently engage with AI-driven outputs.
The firms that will lead in the next decade are those that combine smart automation with informed human judgment, ensuring AI enhances rather than replaces the expertise that define the tax function. A great example from the speakers at the forum was how, once upon a time, people thought calculators would replace accountants, and later, that Excel would make the role obsolete. Instead, these tools elevated the profession. AI is set to do the same.
Generative AI
To maximise the effectiveness and reliability of generative AI in tax processes:
- Clearly define tasks: Specify the inputs and clearly identify the acceptance criteria and constraints. Clearly defined tasks increase the accuracy of AI outputs.
- Consider multi-agent chains: Adopting multi-agent workflows with specialised agents for each function i.e. drafting, fact-checking, will ensure the process is scalable.
- Standardise human hand-offs: Provide all the relevant information, attachments and complete data to the AI upfront. This will reduce errors and improve the process efficiency.
AI governance and security
As AI becomes increasingly embedded in tax functions, robust governance and security frameworks are essential. Businesses must ensure that AI systems operate within clearly defined policies, maintain data privacy and comply with regulatory requirements. Governance includes monitoring AI outputs for accuracy, managing risks of bias or errors, and maintaining transparency in decision-making processes. Security measures, meanwhile, protect sensitive financial data from breaches and misuse, safeguarding both the organisation and its stakeholders. By establishing strong governance and security practices, organisations can confidently leverage AI while mitigating operational, legal and reputational risks.
The future of AI in tax
AI adoption in tax is rapidly accelerating, with authorities such as HMRC and Estonia’s ETCB already using AI for fraud detection, risk scoring and improving service efficiency.
AI is transforming how tax teams work, enhancing human productivity rather than replacing it. The future of tax will see a scenario where humans are assisted by AI, using it to make smarter, faster and more informed decisions.
Over 50% of ITR AI in Tax attendees highlighted that efficiency improvement is the biggest driver for implementing AI, while 54% said process automation will have the most immediate impact on their tax function.
Conclusion
AI is no longer a distant possibility for tax functions – it is an emerging reality reshaping workflows, decision-making and compliance processes. The ITR AI in Tax Forum highlighted that successful AI adoption combines technology with human expertise, clear strategies and rigorous quality controls. By embracing early adoption, structured workflows, generative AI, robust governance and security measures, tax teams can unlock significant efficiency gains while maintaining accuracy and compliance. Organisations that act now will be best positioned to thrive in an AI-driven tax landscape, turning innovation into measurable business value.
About Fintua
At Fintua, our solution Comply automates the end-to-end VAT compliance process across 180+ jurisdictions. This helps save hours of manual effort, boost accuracy and efficiency across the VAT process, enabling your team to focus on higher-value, strategic initiatives.
Subscribe to our newsletter
Stay informed about the latest VAT news, trends and topics from around the globe with our monthly newsletter. Each month, we deliver insightful updates straight to your inbox, helping you stay ahead of the curve.
Authors
Mark Cliffe
Mark is a senior engineering leader with over 25 years of experience delivering scalable, high-performance software in regulated, data-sensitive industries. As Chief Engineer at Fintua, he leads the modernisation of enterprise platforms to enable international growth and seamless client onboarding. Mark has a proven track record in building systems that support high-volume transactions with near-limitless scalability. He combines deep technical expertise with a strategic mindset, leveraging cutting-edge technologies to deliver secure, reliable, and future-ready solutions. Known for developing high-performing global teams, Mark ensures strong governance, continuous innovation, and alignment between technology and business success.
