Tax professionals are increasingly turning to artificial intelligence to speed research, analyze documents and even draft returns — but federal privacy rules haven’t kept pace. The result: a growing gray area around whether and when preparers must tell clients that an AI tool handled their tax data, and what protections govern that sharing.
Old rules, new technology
Federal law designed to protect taxpayer data — notably Section 7216 of the Internal Revenue Code — generally bars tax preparers from using or disclosing a client’s information for purposes beyond filing a return without explicit consent. The statute’s last detailed IRS guidance, however, dates back more than a decade, before generative AI became common in offices.
In June, the IRS issued its first round of instructions aimed at AI use by tax professionals. Those recommendations require practitioners to review and verify any AI-produced work and to reflect efficiency gains in billing. But the agency stopped short of mandating that preparers disclose the use of generative AI to clients.
That omission matters because the line between traditional tax software and AI-powered platforms is not always clear. Some systems only process data for a single return; others may learn from inputs in ways that could expose client information to broader model training.
How widespread is AI in tax work?
Surveys of the industry show rapid uptake. A 2024 Thomson Reuters Institute study of several hundred tax and accounting professionals found roughly one in four had used public, open-source generative AI tools on the job, while a much smaller share — under 10% — had turned to proprietary tax-specific generative systems.
More recent polling conducted by an AI tax research firm and an industry group indicates even broader, regular use. In that study, about 60% of respondents said they now use AI for tax research on a weekly basis — a sharp increase from roughly one-third the prior year.
- Research: Frequently used for pulling case law and guidance (about 60% weekly use).
- Advisory work: Employed for client advisory projects by roughly 44% of respondents.
- Tax planning: Used in planning strategies by around 40%.
- Compliance research: Applied by roughly 39% for regulatory and compliance queries.
- Document analysis and drafting: Reported by about 36% and 35%, respectively.
Industry concerns and differing views
Accounting groups and practitioners say the profession needs clearer direction. The American Institute of Certified Public Accountants has asked the IRS for updated guidance that specifically addresses how new technologies — including AI — should be used in tax preparation and what disclosures, if any, are required.
Some preparers argue AI is distinct from ordinary tax software. Joshua Youngblood, an IRS-enrolled agent and owner of an AI tax-research product, notes that many AI tools can make judgment calls or provide recommendations, which raises different privacy and professional-responsibility questions than software that simply computes a return.
At the same time, software vendors contend that when an AI platform is used solely to prepare a client’s return and the data is protected and not repurposed, the tool functions similarly to conventional third-party processing. That view emphasizes the continuing responsibility of the preparer to review results and remain accountable for filings.
Practical steps clients should take
Until regulators clarify the rules, experts suggest consumers be proactive. Ask these questions before you sign off on a preparer’s work:
- Do you use AI in preparing returns or in your firm’s workflow?
- Will my data be used to train models or shared with third parties?
- What specific guardrails and oversight processes do you have in place to protect client information?
- Will AI-assisted work be reflected in my bill, and how will you document review and verification?
- Can I get a written disclosure or a signed permission form describing how my information will be handled?
Experts also note a procedural distinction: for non-individual filings, disclosures under Section 7216 can be included in engagement letters or memoranda of understanding. For individual returns, the law requires a separate signed disclosure.
Given the uncertainty, many in the profession advise caution. Henry Grzes, who leads tax practice and ethics work at the AICPA, recommends preparers obtain signed client consent when AI plays a role in preparing returns — a conservative approach intended to avoid later exposure if regulators tighten rules.
The stakes are real. Willful violations of Section 7216 can carry criminal penalties — fines up to $1,000, imprisonment for up to one year, or both — in addition to other civil or administrative sanctions depending on the circumstances.
What’s next
Industry groups have already asked the IRS for clearer, modernized directions that address AI’s nuances, including how training and closed systems should be treated. For taxpayers, the immediate takeaway is simple: treat AI use as a material part of the engagement and get clear, written answers about how your data will be handled and protected.
As AI tools become routine in tax offices, both clients and preparers will need to navigate evolving standards. Without clearer federal guidance, transparency and signed consent remain the most reliable way to reduce risk and preserve trust.
Similar Posts
- AI tax help from Grok raises red flags: experts urge caution
- AI threatening financial advisors: MIT professor points to a major obstacle
- Tax deadline hits: millions risk up to $680 in fines after Jan. 31 cutoff
- Get $150 Now: File Your Claim on This Platform Before Sept. 22 Deadline!
- IRS refunds delayed for 830,000+ filers after paper check processing change

Jordan Keller specializes in analyzing the US financial markets. With concrete recommendations, he helps you secure and boost your investments by providing strategies that adapt to market fluctuations.