Tax professionals are beginning to feel the effects of artificial intelligence and experts say the biggest changes are still ahead. Ahead of the ITR AI in Tax Forum in London, Tim Zech from WTS Germany and Jeff Soar from WTS UK shared their thoughts on what the next few years will look like.
Both experts agree that AI will soon become a normal part of daily tax work, not a special project. Systems will handle the first steps of routine tasks such as drafting replies to tax authorities, preparing standard documents, or checking data for mistakes. The human professional will then review the work, use their judgement, and give final approval.
Speed will also improve dramatically. Research that currently takes several days will be completed in hours. This matters because tax authorities are also starting to use AI to identify risks and select cases for investigation. They will expect faster responses from companies as a result.
Jeff Soar highlighted three areas where AI will quickly move from experiment to everyday use:
Research and writing support. AI tools will read legislation and case law, then produce a first draft of a memo or legal response. The tax professional reviews and finalises it, but the hardest part is removed.
Document review. AI will scan contracts, invoices, and financial records to flag potential tax problems early. This is particularly valuable during audits and business acquisitions.
Scenario planning. AI will model the tax impact of business decisions, such as moving operations to another country or changing a supply chain, before those decisions are made.
Tim Zech added a fourth area: knowledge management. Many companies have years of valuable documents like memos, rulings, internal notes stored in shared folders or in people's memories. AI can make this knowledge searchable and accessible, which helps teams work more consistently and trains new staff more effectively.
Tax authorities will use AI to compare company declarations against other data sources, such as payroll records and financial statements. This means audits will become more targeted and specific, rather than broad and random. Companies need to keep their tax data clean, well-organised, and easy to explain.
For tax advisers, client expectations will rise. Clients will want advisers who not only understand tax law but can also offer AI-powered services with faster analysis, smarter tools, and help designing AI processes within the client's own organisation.
One common misunderstanding is that AI will replace tax professionals. In reality, it will replace certain tasks, mainly repetitive ones. Tax work involves ambiguity, negotiation, and ethical judgement, which remain human responsibilities. The greater danger, according to Jeff Soar, is complacency: trusting AI output without questioning it simply because it sounds confident.
Tim Zech pointed to governance as another major risk. Many organisations are still using general AI tools with sensitive data and without proper rules in place. In the coming years, regulators will pay closer attention to how AI is used in financial decision-making. Companies will need to show that their AI-assisted conclusions are explainable and defensible.
Rather than trying to apply AI everywhere at once, the experts recommend a focused approach:
Start by identifying real problems, such as slow audit responses or inconsistent advice across different countries.
Choose three to five specific use cases and do them well before expanding.
Invest in organising and structuring your tax data and documents, because AI is only as good as the information it works with.
Set clear rules early about what AI can and cannot do, and how outputs are checked.
The strategy should also be built for the long term. AI models will improve, regulations will change, and internal expectations will grow. A good AI strategy is not a one-time project, it is an ongoing capability.
The experts identified three types of roles that will matter most going forward. First, all tax professionals will need to be comfortable using AI, knowing how to give it good instructions, review its outputs critically, and recognise when something is wrong. Second, a smaller group of "tax technologists" will bridge the gap between tax knowledge and data systems. Third, specialists in governance will ensure AI is used responsibly and in line with regulations.
Leadership will also need to change. Heads of tax will need to set a clear vision, manage their teams through significant change, and work closely with senior management on AI-related decisions.
