A senior accountant pastes a client's trial balance into a
free chatbot to find a reconciling difference. A seasonal preparer uploads a
stack of K-1s to an AI tool that promises to pull out the numbers. Both save an
hour. Neither checked where the data went.
That's why an AI policy for accounting firms belongs on the
next partner meeting agenda. Your staff are already using AI. What matters now
is whether client tax and financial data stays inside tools the firm controls.
This guide is for firm owners and partners at small and
midsize certified public accountant (CPA) and tax practices, including firms
here in Chicago. It covers why the risk is specific to accounting, what the
policy should say, how to review AI-assisted work, how to vet vendors, and what
good looks like. (This is general information. Talk to your attorney or tax
counsel about your firm's specific obligations.)
Why does an AI policy for accounting firms matter now?
An accounting firm needs an AI policy now because staff are
already using AI, often on personal accounts, with data that federal rules and
professional standards require you to protect. Without written rules, you can't
control which tools see client returns, who checks the output, or what happens
when something goes wrong.
The gap between use and oversight is wide. Verizon's 2026 Data Breach Investigations Report found
that 45% of employees are now regular AI users on corporate devices, up from
15%, and 67% of users reach AI tools through non-corporate accounts on those
devices. Those personal accounts sit outside your firm's controls.
Professional firms are already a top target. The same report
counted 2,558 confirmed breaches in the Professional services sector, which
includes accounting firms, with credentials compromised in 31%. Every unmanaged
AI account is one more login tied to client data.
The cost shows up later. IBM's 2025 Cost of a Data Breach research found that
63% of the organizations studied had no AI governance policies, and high levels
of shadow AI (unapproved AI tools) added USD 670,000 to the average breach
cost.
Can CPAs put client tax data into ChatGPT?
CPAs should keep client tax data out of consumer AI tools
until the firm has reviewed the tool's terms, security, and data handling, and
confirmed whether client consent is needed. Tax return information carries
federal disclosure rules on top of American Institute of Certified Public
Accountants (AICPA) and state board confidentiality duties, so the bar sits
higher than for most businesses.
3 sets of rules matter here:
·
Section
7216. As the AICPA's *The Tax Adviser* explains, Internal Revenue Code Section 7216
is a criminal provision that bars tax return preparers from knowingly or
recklessly disclosing or using tax return information. Disclosure means
"the act of making tax return information known to any person in any
manner." Some disclosures to service providers that help prepare a return
don't need consent. Others need the taxpayer's consent before the disclosure
happens.
·
Professional
confidentiality. The same article notes that client information also falls
under confidentiality requirements from the AICPA, state boards of accountancy,
and privacy laws, plus the terms of your engagement letters.
·
The
Federal Trade Commission (FTC) Safeguards Rule. The FTC lists tax
preparation firms among the financial institutions the rule covers. Its compliance guide says firms using third-party
apps to store, access, or transmit customer information must have procedures
for evaluating their security, and must monitor their service providers.
Where a specific AI tool falls under Section 7216 depends on
the tool, the contract, and how your team uses it. Get that answer from tax
counsel before client return data goes in, and write it into your policy.
What should an AI policy for accounting firms include?
An AI policy for accounting firms should list approved and
prohibited tools, define which client data can never go into unapproved AI,
require a reviewer to verify AI-assisted workpapers and returns, set rules for
client consent and disclosure, and explain how to request a tool or report a
mistake. Most firms can cover it in 3 to 5 pages.
The core sections:
·
Approved
tools. A short list of business-grade tools the firm has reviewed, and a
clear ban on personal accounts for client work.
·
Data
rules. What never goes into an unapproved tool: Social Security numbers,
returns and source documents, bank and payroll data, workpapers, and anything
covered by an engagement letter's confidentiality terms.
·
Consent
and disclosure. When the firm needs Section 7216 consent, how it's
documented, and how you tell clients about AI use.
·
Human
review. No AI output reaches a return, a workpaper, a financial statement,
or a client email until a qualified reviewer checks it against the source
documents.
·
Vendor
diligence. Who approves a new AI tool, and what the vendor must show before
client data touches it.
·
Requests
and incidents. One path to request a tool, and one person to call if client
data goes into the wrong place.
·
Training
and sign-off. Everyone completes short training and signs the policy every
year.
The AICPA also offers its members a Small Firm Generative AI Policy Template,
dated April 2026. It's a useful starting point for the professional-standards
side of your policy.
What Illinois law means for Chicago accounting firms
If you employ people in Illinois, an amendment to the
Illinois Human Rights Act (HB 3773) took effect January 1, 2026. It requires
employers to notify employees when AI is used in decisions such as hiring,
promotion, and discipline, and it bars AI use that has a discriminatory effect.
If your firm screens seasonal hires with AI tools, your policy should say so
and name who signs off.
How should firms review AI-generated workpapers and returns?
Firms should treat AI output like work from a new staff
accountant: fast, often useful, and always reviewed. A qualified reviewer ties
every AI-extracted number to the source document, verifies every cited
authority, and signs off before anything reaches a return or a client. Review
standards should tighten as AI increases volume.
Former Internal Revenue Service (IRS) Commissioner Danny
Werfel laid out the risks in an August 2026 risk framework for AI in tax preparation
published in *The Tax Adviser*. His register for tax preparers lists 15 risks
in 5 categories. Several hit accounting firms directly:
·
Hallucinated
advice: a confident tax position that's wrong or legally unsupported.
·
Client
data leakage: client information exposed to or retained by a vendor in ways
the client never consented to.
·
Incorrect
document extraction: an AI tool misreads a W-2, 1099, or K-1, and the error
flows into the return.
·
Control
dilution: review gets thinner as AI pushes more returns through the same
process.
His recommendation matches what we tell partners. Start with
a limited scope, like summarizing documents, and expand to drafting tax
positions only after your review steps are tested.
Practical review rules:
1. Tie
every AI-extracted figure back to the source document.
2. Check
every code section, regulation, or ruling the AI cites.
3. Keep
the reviewer's sign-off in the workpapers.
4. Track
which engagements used AI so you can spot patterns after busy season.
How do you vet an AI vendor before it touches client data?
Vet an AI vendor by confirming in writing that it won't
train on your data, how long it keeps inputs, who at the vendor can access
them, which security audits it holds, and whether it will sign a data
processing agreement. If a vendor can't answer clearly, keep client data out of
the tool.
Questions to ask every vendor:
·
Do you use our inputs to train models? Where
does the contract say so?
·
How long do you retain prompts and files, and
can we shorten that?
·
Which security audits or certifications do you
hold?
·
Can administrators control which users and
features have access, and is there an audit log?
·
Will you sign a data processing agreement?
This is why we offer Managed Framework AI to our partners.
It runs on an enterprise platform that's System and Organization Controls (SOC)
2 Type I, SOC 2 Type II, and SOC 3 certified. Firm data is encrypted, isolated
from other organizations, and never used to train AI models. Administrators
control which roles can use which models and features, with audit trails, and
data processing agreements are available. Some models carry a 30-day retention
exception, and administrators can restrict those models by role, which matters
for firms handling tax return information.
What does good AI governance look like at an accounting firm?
Good AI governance at an accounting firm means partners own
the policy, staff use one approved platform with confidence, client tax data
stays inside controlled systems, and every AI-assisted number gets reviewed. It
also means the policy gets a fresh look after each tax season as tools change.
At Framework IT, responsible AI guidance is part of every
managed services engagement, and we commit to helping each partner identify at
least 1 real-world AI use case. Your virtual Chief Information Officer (vCIO)
covers AI strategy alongside security and budgeting.
For firms that want help putting AI to work:
·
Training:
our AI Champion Certification takes about 90 minutes.
·
Monthly
check-ins: partners in our Managed Framework AI adoption program meet with
us monthly for an AI Strategic Business Review.
·
Resources:
every Managed Framework AI partner gets the Framework AI Resources Hub from day
one, with training videos, getting-started guides, and industry playbooks,
including one for accounting.
AI governance also ties to your security and insurance
posture. Our guide to cyber insurance for accounting firms covers
the controls carriers and the FTC expect. For the general version of this
guide, see writing an AI policy for a small business. You
can also learn about our AI services and how we support accounting firms in Chicago.
Frequently Asked Questions
Can accountants use ChatGPT for client work?
Keep client tax and financial data out of personal or free
accounts. Consumer tools may retain or train on inputs and give the firm no
visibility. Use an approved business platform with written data terms, and
confirm any Section 7216 consent needs with tax counsel.
Does Section 7216 apply to AI tools?
It can. Section 7216 covers making tax return information
known "to any person in any manner," and some disclosures need the
taxpayer's consent first. Whether a given AI tool needs consent depends on the
tool and how it's used, so get a written answer from tax counsel.
Is there an AI policy template for CPA firms?
Yes. The AICPA offers a Small Firm Generative AI Policy
Template to its members. Use it as a starting point, then add your firm's
approved tools, data rules, review steps, and vendor requirements.
Do Chicago accounting firms have to follow Illinois AI rules?
If you employ people in Illinois, yes, for employment
decisions. Since January 1, 2026, HB 3773 requires notice when AI is used in
decisions like hiring and promotion. Illinois' Personal Information Protection Act also
requires reasonable security measures for Illinois residents' personal
information.
The Bottom Line
Your staff will use AI whether or not you write rules. A
clear AI policy for accounting firms protects client tax data, keeps review
standards intact, and lets your team save time without guessing where the lines
are.
We guide Chicagoland firms on AI policy, and through Managed
Framework AI we help secure the tools and train the team before the next busy
season.
Book an AI readiness conversation with
our team to see where your firm stands and what to fix first.
Framework IT | 700 N. Sacramento Blvd., Suite 101, Chicago,
IL 60612 | (312) 564-5446
About the Author
Adam Barney is President and Managing Partner of Framework
IT, a premier managed IT and telecommunications firm based in Chicago. With
more than 15 years of executive experience in managed services and
telecommunications, Adam leads with a core philosophy that technology should be
user-friendly and approachable, empowering businesses to thrive in their
respective industries.
Since assuming the presidency in January 2020, Adam has led
a team of over 40 professionals spanning sales, information technology,
operations, marketing, human resources, and fulfillment. Under his leadership,
Framework IT remains committed to its unique inverted-pyramid approach, which
ensures clients' needs and aspirations are always the company's top priority.
Founded in 2008, Framework IT specializes in IT support,
strategy, and cybersecurity for small and mid-sized businesses nationwide. The
company's 30+ engineers act as an extension of client businesses, proactively
preventing IT issues so teams have more time to focus on what truly matters.
During his career, Adam has consulted over 1,000 companies, helping them
transform and streamline their technology infrastructure.
In recent years, Adam has spearheaded the adoption of
artificial intelligence in Framework's internal operations and service
delivery, positioning the company at the forefront of AI-driven IT management.
He has pioneered the launch of a new service offering to help clients implement
AI and AI-based automation in their own businesses, enabling organizations to
unlock new levels of efficiency and competitive advantage. Adam is also a
founding member of The Forge AI Alliance of MSPs, an alliance of managed service
providers working to accelerate the adoption of AI and automation in their own
companies and those of their clients.
Under Adam's leadership, Framework IT has been named to the
Inc. 500 Fastest Growing Private Companies in America twice and the Inc. 5000
list at least five times over the past decade. The company has been ranked one
of the Best & Brightest Places to Work in Chicago for five consecutive
years and one of the Best & Brightest Places to Work in the Nation twice in
the last five years. Framework IT has maintained a BBB complaint-free record
since 2008.
Adam's expertise has positioned him as a sought-after voice
in managed services and business technology. His insights have been featured in
the Harvard Business Review, the Washington Post, and Fox 32 Chicago. Adam has
also served as a speaker and panelist at the Chicago Bar Association AI
Symposium, sharing practical guidance on AI adoption with the legal community.
Adam holds a Bachelor of Science in Finance and Business
Administration from the University of Illinois Urbana-Champaign, where he
graduated Summa Cum Laude. He maintains numerous industry certifications,
including ITIL Foundations and multiple AI and cybersecurity certifications.
Outside of work, Adam is a husband and father of two
daughters, as well as an avid reader with interests spanning leadership,
management, psychology, and technology.
