What TR&W 2026 Highlighted About the Direction of Tax Operations

Coming out of this year’s Tax Reporting & Withholding Conference, one thing felt very clear to me. Tax operations is no longer being viewed as a back-office reporting function. It is increasingly being treated as an operational risk function, and more broadly, as a core part of how firms manage data, processes, and client interactions.

Across sessions covering IRS examinations, documentation, digital asset reporting, operational oversight, state reporting, and AI, a few consistent themes kept coming up. Data quality is becoming critical, regulatory complexity continues to increase, and the ability to rely on manual remediation at scale is quickly disappearing. At the same time, there is a shift toward continuous oversight, with AI starting to move into day-to-day operations rather than being treated as a future concept.

What stood out wasn’t just the volume of regulatory change. It was the growing recognition that most of the real challenges sit within operations.

For many firms, the question is no longer whether they understand their reporting obligations. It is whether their onboarding, documentation, validation, and governance processes are strong enough to support those obligations consistently.

Compliance Is Becoming Continuous

A number of sessions focused on IRS examinations, internal health checks, and operational readiness, and there was a clear direction of travel.

Compliance is moving away from being a periodic, reporting-driven exercise and toward something that needs to be maintained continuously. Firms are increasingly expected to identify issues earlier, maintain clear audit trails, and demonstrate control over their processes at any point in time, not just during filing season.

One of the sessions on internal health checks, which Cyrus Daftary, a member of our advisory board, contributed to, highlighted just how important this has become in practice. There is a growing expectation not only to have controls in place, but to be able to show that those controls are operating effectively and consistently producing the right outcomes.

This is an area where independent validation becomes particularly valuable, especially for internal health checks. Automated systems and AI can help teams quickly test and re-test logic, delivering a level of speed and consistency that is difficult to achieve manually, while ensuring processes are working as intended and producing the right outcomes. This is also something traditional internal audit teams are not always set up to do, as it requires a combination of technical capability and tax-specific expertise. An independent layer that can validate logic and outputs in this way provides a much higher degree of confidence in the overall process.

In practice, that means improving data quality at source and validating data on an ongoing basis, rather than trying to fix issues further downstream.

This shift changes the role of tax operations quite significantly. It becomes less about producing forms at the end of the process, and more about maintaining the integrity of the data and processes that feed into them.

Most Reporting Problems Start Upstream

The conference also reinforced just how much complexity is building across reporting regimes, from CRS 2.0 and CARF through to Form 1099-DA, state reporting differences, and the move from FIRE to IRIS.

But the more interesting point was where problems are actually starting.

In many cases, reporting issues are not caused by the reporting engines themselves. They originate much earlier in the lifecycle, during onboarding, documentation collection, validation, and ongoing data management.

Poor or incomplete data collected at the start is then reused across reporting, withholding, remediation, and audit processes, and the impact compounds over time. What starts as a small gap at onboarding can become a much larger issue when reporting deadlines approach.

This came through quite clearly in discussions around CARF self-certifications. I asked a question on this during one of the panels, and it was something that also came up repeatedly in conversations with clients and partners throughout the conference.

At the moment, there is still no clear industry consensus on whether firms will adopt separate CARF self-certifications or move toward hybrid forms that combine CRS and CARF.

What does seem to be emerging is a preference for beneficial owners to use a service provider’s own form templates when opening accounts. At the same time, there are concerns that hybrid forms may not be consistently accepted when documentation is passed between providers, creating additional operational challenges.

It is a good example of how regulatory change quickly turns into practical questions around documentation design, interoperability, and process ownership.

As a result, firms are starting to shift their focus upstream, putting more emphasis on validating data at the point of collection, strengthening onboarding controls, and improving how documentation is governed and maintained over time. Monitoring and remediation are also becoming more structured, with a greater focus on being able to rely on data at any given point, not just at year end.

In effect, regulatory complexity is translating directly into operational complexity, and firms are having to respond accordingly.

AI Is Becoming Part of the Process

Another noticeable shift was how AI is being discussed within tax operations.

The conversation has clearly moved beyond experimentation. AI is now being considered as part of the operational toolkit, particularly in areas like validation, exception identification, monitoring, and workflow support.

This came through clearly in the session on major technologies shaping modern tax operations, where Cyrus stepped in at short notice. It was an interesting discussion, particularly in how firms are approaching adoption in practice.

What stood out is that many firms are not taking a “big bang” approach. Instead, they are incrementally introducing technology into specific parts of their processes. That feels much more realistic, and in many ways, it is becoming the norm.

There was also a useful distinction drawn between different types of AI. At a high level, skills-based AI focuses on performing specific, well-defined tasks such as validating data, extracting information, or applying rules consistently. Agentic AI is more autonomous, capable of making decisions, managing workflows, and completing multi-step processes with less direct human input. Both have clear value, but they are suited to different use cases, and in many tax operations environments, a more controlled, task-focused approach is often the right starting point.

Another important point that came up is that AI is not perfect from the outset. It requires iteration, oversight, and continuous improvement. Using historical data was highlighted as a practical way to both train AI models and test their accuracy. At the same time, it can help identify errors or inconsistencies in past outputs, creating value beyond just forward-looking automation.

AI is therefore not just about efficiency. It is also a tool for improving quality over time.

But there was also a consistent message throughout the discussion. AI on its own does not solve problems. Without the right controls and governance, it can introduce new risks.

Tax operations is a precision-driven environment, where accuracy and auditability are critical. The firms that are likely to see the most benefit from AI are not those adopting it fastest, but those integrating it carefully into well-controlled processes and focusing on the reliability of outputs.

Hearing Directly from the IRS

One of the highlights of the conference was the IRS session. These are always valuable. While they cannot share everything, the insight they provide into direction of travel is consistently useful. US CARF implementation is still being developed and, although no update was given as to when regulations will be available, the IRS does have this as one of its major focuses.

I also had the opportunity to ask John Sweeney from the IRS about the perceived gaps between FATCA and the newer CARF and CRS 2.0 frameworks, particularly where certain products are now being brought into scope under CRS 2.0 but would not be captured under FATCA. I asked whether there were any plans to update FATCA to address this, and the response was that the IRS is aware of the gaps but there are currently no plans to amend FATCA in the same way, an important point for firms to consider as the regimes continue to evolve at different speeds.

Why This Matters Now

The overall takeaway is that firms are entering a period where operational resilience is becoming increasingly important.

A number of pressures are coming together at once, increased regulatory scrutiny, expanding global reporting obligations, more fragmented jurisdictional requirements, digital asset reporting, growing documentation complexity, and rising expectations around automation.

All of this is forcing a rethink of how tax operations functions are structured.

Activities like validation, documentation quality, governance, and ongoing monitoring are no longer secondary. They are becoming foundational to how firms operate.

Those that invest in getting these areas right early are likely to be in a much stronger position, not just from a compliance perspective, but in terms of scalability, efficiency, and overall resilience.

If you’d like to see how TAINA can simplify and streamline your CARF and CRS compliance journey, we’d be delighted to request a demo.

To stay up to date with our latest insights on tax compliance, automation and regulatory change, sign up for our industry newsletter.