Can AI Prepare a Tax Return Without Human Review?
By AG CPA Co. | Perspectives on Finance & Technology | ~610 words
The question seems to answer itself: of course AI can prepare a tax return without human review. Software has been doing the mechanical work of tax preparation for decades. What changes with current AI is the nature and scope of what can be automated — not just arithmetic, but judgment, categorization, research, and analysis. The question worth asking is not whether AI can do this, but whether it should, and under what conditions.
How Far Automation Has Come
A 2025 report from CPA.com found that some accounting firms have achieved over 80 percent automation of individual return preparation. Platforms now exist that take raw client documents — W-2s, 1099s, partnership K-1s, foreign income statements — and produce review-ready draft returns in minutes, applying multi-year consistency checks, cross-referencing data across schedules, and flagging anomalies that human preparers might miss. EY has noted that AI-powered software can analyze historical tax data to identify patterns and proactively address compliance challenges, including responding to tax authority inquiries in an era of increased transparency requirements.
Wolters Kluwer's 2025 Future Ready Accountant Report found that 40 percent of North American accounting firms already use AI-driven tax research, with 80 percent planning to increase investment. For high-volume, straightforward individual returns — a W-2 and standard deductions, perhaps a simple brokerage account — the case for near-full automation is not unreasonable. The error rates in these contexts may well be lower with AI than with human preparers operating under time pressure.
Where the Limits Are
The picture changes materially with complexity. International tax returns involve interacting treaty positions, foreign tax credit calculations, FBAR and FATCA reporting obligations, and entity classification questions that depend on facts and circumstances that are not always captured in documents. S-corporation advisory work, partnership allocations, and business restructuring transactions involve judgment calls that require understanding a client's broader situation — not just the numbers in front of the system.
The World Financial Review's February 2026 analysis of AI in tax preparation made the observation plainly: automation can be very high, but human oversight remains essential for quality and risk control. Even firms that have invested heavily in AI tools find that the bottleneck shifts rather than disappears. Preparation time falls; review time does not fall at the same rate, because the review must actually be meaningful rather than perfunctory.
The IRS's own posture reinforces this. In January 2024, the agency issued a request for information specifically seeking to understand how AI could be used to identify issues with returns at initial filing — implying that AI-generated returns require external validation, not just internal checking.
The Professional Responsibility Question
For tax professionals, the answer to the article's title question is partly a technical one and partly an ethical one. Circular 230 requires that a practitioner exercise due diligence in preparing and filing returns and other documents relating to IRS matters. The standard does not specify the tools used, but it does specify the professional's responsibility for the output.
A CPA who signs a return prepared entirely by AI and reviewed only nominally has not necessarily met that standard. Conversely, a CPA who uses AI to handle data extraction, initial categorization, and consistency checking — while applying genuine professional judgment to the final product — is arguably meeting the standard more effectively than one working manually under time pressure.
The AI tax preparation tools available in 2026 are best understood as a force multiplier for professional judgment, not a replacement for it. Firms that adopt them thoughtfully — defining clear roles for automation and clear points at which human expertise must engage — are likely to serve clients better, not worse. The risk is not that AI prepares returns. The risk is that the human review layer becomes nominal rather than substantive.
REFERENCES
CPA.com. '2025 AI in Accounting Report.' 2025.
Wolters Kluwer. '2025 Future Ready Accountant Report.' 2025.
EY. 'The Rise of AI in Delivering Smarter Tax Solutions.' January 2026.
World Financial Review. 'AI in Tax Preparation: What to Automate and Where Human Tax Experts Still Matter.' February 2026.
IRS Request for Information on AI in Tax Filing. January 2024.