IRS rolls out introductory AI guidelines for tax practitioners

The Internal Revenue Service introduced preliminary guidelines for tax practitioners using AI that, among other things, call for firms to exercise due diligence when it comes to AI outputs and pass AI-related savings on to their clients. 

Processing Content

This is according to an email the IRS sent to practitioners this morning which defined AI, at a basic level, as the use of machines in a way that mimics human cognitive skills, including judgment, perception, and prioritization, then adding that virtually all professional tax firms use some form of AI, whether they are aware of it or not. While it noted the potential benefits of AI in federal tax practice, it also pointed to limitations such as fabricated outputs, bias and lack of transparency, all of which pose serious ethical and legal risks. 

“As a result, the use of GAI presents concerns involving privacy, confidentiality, and data protection. For example, client privacy can be compromised when data generated for one client is repurposed by the program to respond to an inquiry concerning another client, or data compiled for a particular issue is spilled over into an algorithm and combined with a related tax issue involving a different client. As such, it is incumbent on any tax professional using GAI to carefully review all documents crafted by the technology,” said the IRS. 

The guidelines don’t represent new rules per se so much as clarify how existing rules, specifically Circular 230, apply to the use of AI. 

For example, according to 10.27(a) of Circular 230, “A practitioner may not charge an unconscionable fee in connection with any matter before the Internal Revenue Service.” Based on this, the IRS said that cost savings from AI should be passed on openly, with billing practices that reflect the efficiencies gained from the use of generative AI. Practitioners should not only disclose, in general or specific terms as needed, the AI activities performed, but also fairly credit to the client’s account any cost reductions. They should generally not bill clients for manual labor or time that was not actually spent or double billing for AI-assisted tasks. Depending on the facts, such as a noticeable pattern across clients or the size of the billing differentials, this may constitute a  §10.27 violation. 

Other parts are more connected with governance and due diligence. 

Section 10.22 of Circular 230 says that practitioners must exercise due diligence in preparing or assisting in the preparation of, approving, and filing tax returns, documents, affidavits, and other papers relating to Internal Revenue Service matters; in determining the correctness of oral or written representations made by the practitioner to the Department of the Treasury; and in determining the correctness of oral or written representations made by the practitioner to clients with reference to any matter administered by the Internal Revenue Service.” 

Therefore, said the IRS, when using generative AI practitioners must thoroughly review all AI-created documents and language incorporated into writings before delivery to a client or submission to the IRS. This includes verifying the accuracy of facts, citations, and calculations produced by AI. The IRS noted that practitioners cannot rely solely on AI, saying that human scrutiny and editing are essential to ensure correctness and compliance with IRS expectations.

Along similar lines, section 10.37 requires that practitioners who give written advice concerning one or more federal tax matters subject to the requirements of Section 230 must base such advice on reasonable factual and legal assumptions (including assumptions as to future events); reasonably consider all relevant facts and circumstances that the practitioner knows or reasonably should know; use reasonable efforts to identify and ascertain the facts relevant to written advice on each federal tax matter; not rely upon representations, statements, findings, or agreements (including projections, financial forecasts, or appraisals) of the taxpayer or any other person if reliance on them would be unreasonable; relate applicable law and authorities to facts; and not, in evaluating a federal tax matter, take into account the possibility that a tax return will not be audited or that a matter will not be raised on audit.

As such, the IRS said practitioners cannot rely on GAI projections or representations without verification. For legal documents, citations must be checked and cases read. Financial forecasts, inputs, and formulas need to be confirmed. If the system’s logic is opaque, reliance may be unreasonable under section 10.37. Further, when drafting written advice with generative AI, practitioners must independently authenticate all factual and legal information. Blind reliance on what AI yields, especially when the underlying logic or sources are unclear, may constitute unreasonable reliance. Practitioners should treat the advice as a starting point, subject to thorough review before providing it to clients.

In addition, as the IRS requires the protection of sensitive taxpayer information as per IRC sections 6713 and 7216(a), as well as Section 10.51(a)(15) of Circular 230, practitioners must strictly handle all client data using only secure, enterprise-approved AI. AI systems should be utilized with robust confidentiality safeguards firmly in place. Willful mishandling of taxpayer information through AI may lead to disciplinary actions under Circular 230.

In addition, as per section 10.35, “A practitioner must possess the necessary competence to engage in practice before the Internal Revenue Service. Competent practice requires the appropriate level of knowledge, skill, thoroughness, and preparation necessary for the matter for which the practitioner is engaged” and so the IRS said practitioners must understand both the law and the technology used in their representation of clients before the IRS, including AI systems’ operational mechanics, limitations, and risks. This includes understanding how AI develops content, recognizing the potential for bias or errors, and being able to evaluate whether AI outputs are suitable for use in IRS matters. Lack of technological competence could lead to improper advice or flawed filings.

Finally, the guidelines discuss internal policies. It noted that section 10.36 requires that those overseeing a firm’s tax practice, must take reasonable steps to ensure that the firm has adequate procedures in effect for all members, associates, and employees for purposes of complying with Circular 230. With this in mind, the IRS said firms must deploy internal policies and procedures for compliance with Circular 230 in the AI space. 

For staff this means comprehensive training on use of AI (the risks – technological and other – and the requirements). For the firm’s internal rules, this means established protocols for secure data handling, AI accuracy monitoring, etc. And for contracting with external providers, this means outsourced or third-party AI tools should be vetted. All steps and processes must be documented to show adherence to section 10.36.

Overall, the IRS recommended that people identify, understand, and stay updated on any relevant federal or state-specific laws, regulations, and guidance that pertain to your professional activities; establish secure AI data handling protocols and access controls; document AI usage and verification processes; foster transparency and accountability in all AI practices; prepare clear procedures for handling breaches or errors; provide necessary staff training; vet third parties’ AI offerings when or before purchasing; never upload sensitive data to unsecured sites; treat the written text that AI generates as drafts; and review the resulting documents thoroughly for factual and legal accuracy (e.g., always check citations) and any problematic bias.

“GAI holds promise to improve efficiency and professional services in tax practice. However, ethical obligations of competence, diligence, and confidentiality remain unchanged. By implementing robust use-management strategies and maintaining human supervision, tax professionals can harness AI’s benefits while safeguarding reliability and public trust,” said the IRS.