Why Trust May Be the Most Important AI Challenge for the Tax Ecosystem
The adoption of artificial intelligence into work processes is often described as a race. Companies are racing to adopt it while software providers are racing to build it. As AI becomes the norm within the business sector, the Government is slowly trying to figure out where and when its race should start.
The benefits of AI are often touted as speed and efficiency, which leads to questions. How quickly can AI analyze information? How many hours can it eliminate? How much more productive can organizations become?
For tax agencies and governments at large, the question has significant moral implications: Will AI erode trust? That distinction matters because tax administration depends on public confidence. Taxpayers may not agree with every law, audit finding, or filing requirement, but they generally need to believe the system is consistent and accountable. When that belief is called into question, so is the trust that bolsters voluntary compliance.
How should tax professionals think about AI’s growing role in enforcement and compliance? Join the Sales Tax Institute’s upcoming webinar, AI in State Tax: New Tools, Familiar Risks, for an expert discussion on the opportunities and possibilities that AI introduces for the future of tax administration.
As AI becomes more common in both the public and private sectors, tax professionals are considering what AI use means for transparency, bias, and accountability. Understanding why trust is such an important part of the conversation begins with recognizing that AI is not arriving in a vacuum. Governments have been using data to support enforcement decisions for decades.
AI May Be New, but Data-Driven Tax Administration Is Not
AI entering government workflows may sound like a dramatic shift, but tax agencies have spent decades using data to spot anomalies and focus enforcement efforts where they believe the risk is greatest. Modern AI initiatives are built on decades of investment in tax technology, automation, and data analytics. Today’s AI-powered tools represent the latest stage in a much longer evolution toward increasingly automated tax processes. Rather than replacing everything that came before it, AI is being layered onto familiar foundations.
That distinction is important because it highlights why today’s conversation feels different. Agencies are not suddenly discovering the value of data or analytical technologies, but evaluating whether a new generation of tools can improve how those capabilities are already used. The difference is that generative AI arrived with a level of public visibility that previous technological advances never experienced.
Yet greater capability often creates greater responsibility. The more influential a tool becomes in the decision-making process, the more questions stakeholders ask about how individuals reach those decisions. The challenge for tax agencies is not simply adopting new technology. It is doing so in a way that strengthens confidence in the systems that taxpayers depend on.
Why AI Creates a Different Kind of Public Conversation
Many emerging technologies spend years operating largely in the background before the average person encounters them. Generative AI has followed a different path. For the first time in decades, everyday consumers, businesses, regulators, and elected officials are all interacting with the same technology at roughly the same time.
According to a Federal Reserve analysis of generative AI adoption, roughly 40% of working-age Americans reported using generative AI, a level of public exposure that most emerging technologies never achieve so quickly. That widespread familiarity means AI is no longer confined to technology professionals or innovation teams. It has become part of daily conversations in workplaces, classrooms, and homes. As a result, AI is no longer a niche technology discussion. Instead, it has opened up a dialogue about our responsibility in using it, and how much, or how little, the public is willing to trust this technology.
People already understand enough about AI to have opinions about it. They have read stories about inaccurate outputs and seen examples of bias. Concerns about privacy, intellectual property, and data security are top of mind as businesses experiment with different forms of adoption. For tax agencies, perception matters because trust is one of their most important assets.
A private company may accept a certain amount of experimentation in exchange for speed. Government agencies generally have less flexibility. They must demonstrate fairness, consistency, and accountability while maintaining public confidence that their privacy is maintained. Even a highly effective tool can become problematic if taxpayers believe its conclusions cannot be explained or challenged.
This is why discussions about AI frequently return to transparency. Decisions that affect taxpayers must be understandable, defensible, and subject to appropriate oversight. Accountability cannot disappear simply because a technology-assisted process was involved, and safeguards are necessary to mitigate bias, errors, and other unintended consequences. This is why the adoption rate of AI within the government looks much different than in a public company. Ryan Minnick, Federation of Tax Administrators explains that “States are currently looking at two primary paths for the adoption of AI. The first is how to enable end users experience using self-directed AI tools like chat bots. The second is identifying workflows that be enhanced using AI such as predrafting communications.”
These challenges do not have simple solutions. More importantly, they are not purely technical issues. They are governance issues.
Regulation Is Attempting to Catch Up
The challenge becomes even more complex when regulation enters the picture. The European Union’s AI Act has established one of the most comprehensive regulatory frameworks to date, taking a risk-based approach to AI governance. At the same time, states across the United States continue to introduce their own requirements and disclosure obligations.
Rather than creating a single set of expectations, the result is an increasingly fragmented landscape. The same AI-enabled process may be subject to different requirements depending on where data originates, where customers are located, and how the technology is used.
Organizations are increasingly balancing innovation, compliance, transparency, and risk management at the same time. Tax agencies will face many of these same pressures as they evaluate how AI fits within existing frameworks while maintaining the confidence of taxpayers.
The Future Will Be Defined by Governance, Not Technology
Throughout history, new technologies have promised greater efficiency, and AI is no exception. What makes this moment different is that government agencies are being asked to evaluate a rapidly advancing technology while operating within institutions intentionally designed to move slowly.
While the popular narrative around AI focuses on what is possible, the more consequential discussion may be what is acceptable. The future of AI in tax administration will depend on transparency, oversight, accountability, and the ability to balance efficiency with privacy and taxpayer protections. Policies surrounding those issues may ultimately have a greater influence on adoption than any individual technological breakthrough.
Tax agencies are responsible not only for administering tax laws but also for maintaining confidence in the systems that enforce them. As AI expands into more government functions, success will depend on more than improved efficiency. It will depend on whether taxpayers continue to view those systems as fair, transparent, and worthy of trust.
In many ways, that is the central question facing tax administration today. Not whether AI can make governments faster, but whether governments can embrace AI without compromising the trust that makes effective tax administration possible in the first place.
The decisions made today about accountability, transparency, and oversight are likely to shape the future of tax administration far more than any single AI model ever could.
Want to explore where AI, tax administration, and public trust may be headed next? Register today for AI in State Tax: New Tools, Familiar Risks. Beyond the educational content, the session will feature a fireside chat-style conversation with Diane Yetter of the Sales Tax Institute and Ryan Minnick of the Federation of Tax Administrators (FTA) examining the broader implications of AI in government, the direction regulation may take, and what tax professionals should be watching as this technology continues to evolve.