For many accounting firms, the first conversation about artificial intelligence was about efficiency.
Could AI help staff research faster? Draft routine communications? Automate repetitive work? Reduce the hours required for bookkeeping, tax preparation or reporting?
A more uncomfortable question is beginning to sit behind those discussions.
What happens when the client starts using the same technology?
A business owner who once called an accountant for a quick analysis, a reporting question or help interpreting financial information can now ask an AI system first. The answer may not always be right, and complex judgment still requires experience and accountability, but the behavior itself matters.
AI does not have to replace a CPA completely to change the relationship. It only has to remove enough of the routine interactions that once kept the accountant close to the client.
For regional and boutique firms, that may be the more immediate risk: not a sudden disappearance of the profession, but a gradual loss of relevance as technology moves closer to the client.
The Client Relationship Is the Real Asset
Regional CPA firms have traditionally competed with an advantage that is difficult to reproduce quickly: trust.
A CPA who has advised the same owner for ten or twenty years may understand the business far beyond the financial statements. The accountant may know where margins are under pressure, which employees hold critical knowledge, where cash gets tight, which systems create recurring problems, and how management actually makes decisions.
That familiarity has value because business problems rarely arrive in neat categories.
A discussion that begins with profitability can turn into a question about staffing. A reporting problem can expose weak data. A budgeting discussion can reveal an inefficient workflow. A tax conversation can lead to a broader question about systems, automation, or growth.
Historically, the CPA was often one of the first calls because the relationship already existed.
The concern is what happens when AI begins answering more of the first questions.
If clients become accustomed to going first to software, an AI platform, or an outside technology provider, the CPA may lose more than a piece of billable work. The firm can lose visibility into the problems that create future advisory opportunities.
AI Can Change the Economics Before It Replaces Anyone
The pressure is also coming from the other direction.
Large accounting and professional-services firms have spent years investing in technology that makes traditional work more efficient. AI is increasingly being applied to areas such as tax, bookkeeping, payroll, research, document review and internal workflow.
When the labor required to deliver a service falls, the economics change.
A client that may once have been too small for a national firm can become more attractive when technology allows the work to be performed faster and with fewer people. Private equity-backed professional-services groups and technology-enabled providers can make similar investments.
That means the regional firm is no longer competing only with another accounting practice across town.
It may be competing with a larger firm that can now serve smaller clients economically, an AI-enabled provider with lower delivery costs, or software that allows the client to handle portions of the work internally.
None of those forces requires AI to become a perfect substitute for an accountant.
They only require it to make certain parts of the relationship easier to unbundle.
Being the Trusted Adviser May Require a Broader Answer
At the same time, clients are asking new questions.
Which AI tools should the company use? Which processes are worth automating? Can employees safely put company information into public generative AI systems? What happens when an AI agent receives access to internal software? How should data be governed? What does a useful AI project actually look like?
Those questions often land with CPAs because CPAs are already trusted advisers.
The problem is that many regional firms do not have AI engineers, cybersecurity specialists, implementation teams, or technical salespeople sitting in-house.
The traditional answer has been to refer the work elsewhere.
That can be the right decision, but there is a strategic consequence when the same pattern repeats. The outside provider begins learning the business. It gets closer to management. It becomes involved in operating decisions, software choices, data and strategy.
Eventually, the provider may become the person the client calls first when a new problem appears.
For an accounting firm worried about losing clients to AI, that is an important distinction. The risk is not only that software performs accounting work. It is that the center of the advisory relationship moves toward whoever is helping the client navigate technology.
What the Largest Firms Learned Earlier
Oakspring Labs co-founder Val Kharkover says one of the lessons he took from working in a Big Four environment was that technology could be used not only to improve delivery, but also to understand existing clients more deeply.
Kharkover participated in a client-expansion initiative that combined Salesforce with AI-assisted prioritization to identify likely opportunities across an existing client base.
The important part of that example is not the analysis itself.
It is where the opportunity came from.
The firm did not need to begin with an entirely new market. It already had relationships. Technology helped surface where those relationships could expand.
That same logic may be increasingly relevant for regional CPA firms.
Their clients are already spending money on software, automation, cybersecurity, data and AI. The question is whether those dollars and the strategic conversations surrounding them remain close to the CPA firm or move entirely to outside providers.
A Third Option Between Building and Referring
Oakspring Labs was founded in 2025 by former Big Four Tax and Advisory professionals around that problem.
Its Oakspring Advisor Partnership is designed for mid-tier, regional and boutique CPA firms that want to add AI and technology advisory capabilities without building a complete technical organization themselves.
The structure is deliberately behind the CPA relationship.
The accounting firm remains in front of the client and decides which relationships to pursue. Oakspring provides capabilities such as opportunity analysis, sales support, AI engineering, implementation and ongoing technical delivery behind the scenes.
The objective is not to turn the CPA firm into a software company.
It is to give the firm a way to stay involved when clients begin asking questions that extend beyond traditional accounting work.
That distinction becomes more important if AI continues to compress routine services. A firm that can participate in technology decisions, identify operational opportunities and help clients implement useful systems has more ways to remain valuable even as individual tasks become automated.
Start With the Business Problem, Not the AI Tool
There is another complication in the current AI market: companies are being offered an enormous number of products before they have clearly defined the problem they are trying to solve.
A new model or application appears, and management begins searching for a use case.
That approach can produce expensive demonstrations with little financial impact.
Oakspring says its process runs in the opposite direction. The starting questions are operational and economic.
Where is money being lost? Where are employees spending unnecessary time? Which process creates a bottleneck? What information is unreliable or difficult to access? Where could better technology increase revenue, reduce cost or create capacity?
Only then does the technology question come next.
Sometimes the answer may involve AI. Sometimes a conventional software integration, workflow redesign or better data structure may be more useful.
For the CPA, this kind of conversation is familiar territory because it begins with the economics of the business rather than the novelty of the technology.
AI Is Also Creating a Security Problem
As AI becomes more useful, it generally needs access to more information.
That creates a second issue for firms trying to advise clients through the transition.
Employees are experimenting with public AI tools. Low-code platforms are allowing non-engineers to build internal applications. AI agents are being connected to company systems. More information is moving between platforms, models and vendors.
In many organizations, experimentation is moving faster than governance.
Beginning September 1, 2026, Oakspring Labs plans to expand the Oakspring Advisor Partnership with broader cybersecurity and data-governance capabilities. Myron Kharkover, co-founder and head of technology at Oakspring, says the two areas are becoming increasingly difficult to separate because an AI system that can access useful company information also creates questions around permissions, credentials, sensitive data, monitoring and accountability.
For CPA firms, that may create another opportunity to remain close to the client rather than referring every technology-related issue away.
The Larger Question for CPA Firms
The most important question may not be whether AI will replace CPAs as a profession.
That framing is probably too simple.
A more immediate question is which parts of the traditional client relationship AI will make easier to replace, automate or move elsewhere, and what firms will do with the space that remains.
Trust, judgment and knowledge of the client are still substantial advantages. But those advantages become more valuable when they are connected to the problems clients are trying to solve now.
Regional firms do not necessarily need the technical headcount of the largest accounting organizations to compete. They do, however, need a credible answer when a client asks what AI means for the business.
If the answer is consistently somewhere else, the relationship may eventually follow.
