As talent shortages continue to challenge accounting firms, especially midsized and large ones, leaders are increasingly turning to automation and artificial intelligence to not only fill gaps but to expand capacity without growing headcount — something that is already having dramatic effects that could signal a new era for the profession.
Accounting Today's own Year Ahead survey found that a significant proportion of firms are still experiencing staffing issues: for large firms, recruiting and retaining good employees was the No. 1 issue they're facing this year, and 27% of midsized firms said the same thing. Many of the employees they currently have are also feeling overworked and demoralized, as 27% of both large and midsized firms also listed "burnout" as among their top challenges. Eva Mrazikova, senior director of product marketing with accounting practice management solutions provider IRIS, said that these challenges "will absolutely persist into 2026 and beyond."
"With over 300,000 accountants having left the profession since 2020 and three-quarters of CPAs approaching retirement age, this isn't a temporary squeeze — it's a structural shift that demands new thinking," she said.
While most firms are seeking to address this the old-fashioned way — hiring more staff, both permanent and seasonal — automation is not far behind. The Year Ahead survey found that 46% of firms intend to hire more full-time staff and 45% said they plan to hire more seasonal staff. Yet more than a third, 35%, also said they anticipate automating processes using AI, and 23% said they will automate processes with more classic solutions.
Lisa Huang, senior vice president of product for small-business accounting platform Xero, said that doing so can serve to lower the per-client cost, allowing for expanded capacity for higher-value work: "The industry continues to face a massive talent shortage against rising demand. AI lowers the 'cost to serve' per client by handling the low-value manual work, which unlocks the capacity for firms to say 'Yes' to clients they previously had to turn away. In 2026, successful firms will leverage this efficiency to rebalance their focus, moving away from hourly billing toward high-value advisory that drives growth for the firm. It's about leveraging technology to do more, not just to spend less," she said.
Agents, integrations and data flows
This has already been happening for some time, with many routine, time-consuming processes now mostly automated. Software solutions can auto-populate forms, categorize and sort transactions, reconcile journal entries, and much more. As technology grows more sophisticated, it is unlikely this trend will reverse anytime soon. Experts believe that this year will see even more automation, especially of the arduous prep work that has to be done before the accounting can even start. Sourcing documents, cleaning data, doing research, and other pre-accounting tasks are predicted to be automated by the end of this year.
"The area that will see the most dramatic reduction in human involvement is pre-processing — everything associated with getting information into systems in the first place. That includes sourcing documents like bills, extracting data from those documents, and categorizing information correctly so it's ready for downstream workflows," said Ariege Misherghi, general manager of the accounting channel for payment solutions provider Bill. "These tasks are necessary, but they're fundamentally mechanical. AI is increasingly capable of handling them faster, more accurately, and at a greater scale than humans ever could. As a result, the manual effort required to turn source documents into structured, usable data is rapidly disappearing."
One of the big differences is the rise of agentic AI. While AI agents have been around for years, recent software advances (at least partially due to generative AI) made them accessible to the general public. While typically AI had been applied to simple, rules-based tasks, semi-autonomous agents will allow the automation of more complex processes, according to Ellen Choi, founder and CEO of accounting-focused AI consultancy Edgefield Group.
"Today, we have working generative and agentic AI in the market that is meaningfully taking over what was once in the exclusive domain of humans, most notably in tax preparation and bookkeeping. However, this only works for very constrained, simpler cases, i.e., tax prep for less-complex 1040s and bookkeeping for single entities. In 2026, technologists, vendors and firms will build on this foundation to apply AI to take on more complex end-to-end use cases, including in tax preparation and bookkeeping," she said.
The effectiveness of these agents has been enabled by another factor: Firms are approaching their AI implementations more deliberately than before. When major advances in artificial intelligence hit the accounting profession a few years ago, they were seen almost as novelties, and leaders had to take active efforts to figure out use cases for the new solutions that were flooding the market.
Compounding things was that many of these products were presented as standalone solutions separate from the firm's main workflows. They might have been used to make existing processes faster and more efficient but, fundamentally, they were the same processes. In contrast, many firms today take a more holistic approach, working AI directly into their workflows not as a bolt-on point solution but as a component woven throughout the process.
Davis Bell, CEO of practice management solutions provider Canopy, said AI will increasingly be operating in the background, almost invisibly. "In 2026, AI will shift from being an optional add-on to a native layer inside the core systems accountants already use: bookkeeping, client management, workflow, document handling, billing and reporting. Instead of opening a separate AI tool, accountants will increasingly experience 'ambient AI:' permission-aware AI that quietly handles summaries, document classification, task creation, data consistency checks, and client follow-up inside the firm's daily workflows. 2026 will be the year AI meaningfully increases firm capacity, realization rates, and partner-level revenue, without increasing partner or admin hours," he said.
Another factor has been platform consolidation. Automation, particularly when it is tied to AI models, is only as good as the data it can access. When data is siloed, systems cannot communicate, which creates gulfs in the automation process that must be cleared manually. But over the past year especially, many organizations have been breaking down those silos in order to facilitate better data flow and, consequently, automation.
"The big shift is platforms talking to each other through AI layers and APIs and not through several humans doing the glue work. You'll see more continuous finance, continuous controls, continuous compliance, and hopefully finally continuous auditing. Orgs will care less about feature lists and more about, 'Does this actually run in the background without making my team miserable?'" said Avani Desai, CEO of Top 50 Firm Schellman.
Vendors that serve the accounting community have been making similar moves with the products they've been offering. Over the past year, many companies have not just added AI features to their offerings, but are reworking their entire system architecture to revolve around agentic AI. At this point, companies generally don't tout AI as the product itself but, rather, the enabling factor that makes the product work. Intuit, Avalara and Oracle are only some of the many solutions providers that restructured their products around AI, while startups like DualEntry have done so out the gate.
"What it means to be AI-native is to have AI embedded in the architecture and infrastructure of the system such that you can go workflow by workflow. You can see where accountants spend most of their time, and then you can go workflow by workflow and help automate those with the use of AI. Rather than just something at the surface level, this is deeply embedded in the actual workflow, in the actual product," said DualEntry co-founder Benedict Dohmen.
Pipeline impacts
Taken together, all of these have worked to dramatically expand what AI agents can do. As they improve, some industry experts — such as Elizabeth Beastrom, president of tax and accounting professionals for Thomson Reuters — say agents will increasingly be seen almost as another worker, and accounting leaders will be managing agents just as surely as they manage humans today. Beastrom predicted that agents will become so reliable that, within five years, firms will have a roughly 1:1 ratio of humans and robots.
"Firms will employ at least one virtual agent for every CPA on staff within the next five years. This 1:1 ratio will crush talent shortages and act as a cost-effective way to bolster productivity and curb burnout. AI agents will handle manual research, data extraction, and routine analysis, culling vital information from trusted sources — like the Tax Code and a firm's own financial documents — to distill key insights and solve specific tax-related problems," she said.
This technological path has led to an unusual shift in how firms budget. On the one hand, by expanding capacity without growing headcount, firms are spending less on personnel; on the other, because of high upfront costs like integrations, cybersecurity and licenses, they're spending more on technology. In other words, capital spending is eclipsing labor spending.
Jack Castonguay, vice president of strategic content development for accounting, finance and AI with accounting education group Surgent, noted that this is a significant reversal from how firms typically have budgeted. Labor used to be a firm's biggest cost center, as well as its most important asset. But with the rise of AI, capital investments, particularly where technology is concerned, are starting to outpace it.
"AI has greatly increased the capex spending at firms already and reduced people costs," said Castonguay. "A firm's most important and most expensive asset was its people. That is changing. Firms are essentially taking any free cash flow they have and investing it right back in AI. I think that is also a big reason we are seeing firms seeking more private equity investments than ever before. And because the firms believe AI will be able to replace up to a third of their existing new hires, they are reducing hiring and not raising wages like they would absent AI," he said.
Choi, from Edgefield Group, made a similar observation: "While AI has added to operating costs (even incremental costs like Copilot licenses add up), firms are realizing meaningful time savings with AI investment, which are translating directly into cost efficiencies. The clearest signal is in workforce planning: Firms are modeling growth while holding headcount steady or maintaining revenue while allowing for natural attrition. Leading firms and operations teams are explicitly baking these efficiency gains into their operating models, reflecting a shift from experimental adoption to deliberate firm planning," she said.
But while AI might address the pipeline problem in the short term, some worry about the long-term impact. Yes, AI is increasingly taking on the routine repetitive tasks that, in the past, were handled by entry-level accountants. But then, how will the new generation of accountants develop the foundational skills necessary for higher-level work if they no longer handle the simple tasks?
Joy Taylor, managing director of consulting for Texas-based advisory firm Alliant, wondered if accountants have no experience doing this work, how well can they understand and evaluate the AI's results? "I do believe that technology and AI will [assist with many of] those transactions, but it will still be critically important for those that inherit the output of those digital results to interpret them and confirm that they are correct, and that is the skill that still requires a great deal of knowledge, careful consideration and an investment in education."
Wesley Hartman, founder of Automata Practice Development, said that training and mentoring will be key to heading off skill erosion. If firms do not do this and simply put things in AI's digital hands, sooner or later they'll encounter succession problems. "I think there needs to be a higher level of coaching and mentoring to elevate someone coming in at entry level to be past the previous expectations," he said. "A senior person with an AI tool will be able to 'manage' agents to do the work of a team of entry-levels, so we need to figure out how to get those entry-levels to seniors faster. Otherwise, in five years, there will be no seniors, and the pipeline will be worse."
Career impacts
Whoever the new hires will be, they will be operating under a different set of expectations than the professionals before them. As AI transforms what it means to be an accountant, it is also changing what firms are looking for in their staff. The ability to tolerate endless, repetitive work without going crazy will be less valuable versus soft skills, technology experience and analytical thinking, according to Cathy Rowe, senior vice president and segment leader at Wolters Kluwer.
"The advisory-first model will continue to accelerate," she said. "With 93% of firms already offering advisory services and nearly half planning to expand these offerings, the accountant of the future will be a data-driven strategist. AI and predictive analytics will become standard tools for delivering proactive, personalized guidance that strengthens client relationships and drives sustainable growth. In 2026, the profession will be defined by a culture of learning, empowerment and adaptability."
This is already leading to the creation of new career paths, and Jenna Shapiro, chief people and culture officer with Top 10 Firm Grant Thornton, believes this will continue over this year. "By 2026, roles like AI compliance officers and finance technologists will emerge as core to the profession. Firms that create room for growth and help people adapt will attract and retain the talent of the future. We're already redesigning career paths and building leadership programs to help our people guide clients through this new era," she said.
Existing roles, meanwhile, will take on new significance. For instance, Aaron Harris, CTO of accounting ERP provider Sage, said the chief technology officer will become even more important as AI grows more influential. "As intelligent systems take on more of the execution layer in finance, someone needs to guide how those systems behave — and yes, I'm biased, but the role best positioned for that is the CTO. We've been preparing for this moment for a long time," he said.
Meanwhile, Dan Priest, the chief AI officer for Big Four firm PwC, said AI could change something much more fundamental to accounting careers in general: a shift from specialists to generalists, something that has been happening in other places, such as the U.S. Navy. Priest's prediction indicates this might be an economywide phenomenon.
"AI could soon end a shift that has marked most of the industrial era — the ever-increasing specialization of work. Agents can increasingly do the specialized tasks that fill the workdays of experienced, mid-tier employees," he said. "In IT, for example, you may no longer need coders specialized in specific languages. Instead, you may want engineers who understand both tech architecture and how to manage and oversee the agents that do know these languages. In finance functions, as agents do tasks like invoice processing, purchase order matching, reconciliation, and anomaly detection, people with general finance skills can focus on growing revenue and expanding margins, engaging with vendors on payment terms, working with sales on dynamic pricing models, and conducting more scenario planning."
With power comes accountability
With artificial intelligence being entrusted with more high-value tasks, the need for effective oversight and governance has grown, especially as risks become more apparent. AI-generated false information (colloquially known as "hallucinations") has been finding its way into court cases, medical imaging, police reports and even government studies. At the same time, workers reported rising irritation from what has been called "AI workslop," low-effort AI-generated work products that serve to shift the burden of the work downstream, requiring the receiver to interpret, correct or redo it, ironically generating even more work than if the creator had just done the task themselves.
Consequently, Micheal Herman, chief digital officer for Top 25 Firm Baker Tilly, expected firms will get increasingly serious about their AI policies. "With AI embedded in critical accounting and audit processes, trust and transparency will become nonnegotiable," he said. "Firms will implement strict AI-use policies, client disclosures, and bias monitoring in the new year. Organizations will prepare to navigate the emerging regulatory landscape with greater precision and will start implementing new governance, controls and audits in order to evidence to the market a responsible AI culture."
Firms likely will also start asking more of their vendors as well. Over the past several years, there have been an astounding number of AI-based products and solutions pitched to accounting professionals at CPA firms, corporate offices and small businesses. Many organizations raced to adopt them as fast as they could, fearful that if they didn't, they would surely be outcompeted by those who did. However, many soon learned that doing so may be more difficult and expensive than they thought, with many failing to generate direct financial ROI.
People like Todd McElhatton, chief operating and financial officer for subscription and monetization platform Zuora, felt that people are no longer impressed with AI unto itself; he anticipates that this year will be about "proof, not potential."
"[Professionals] will start demanding hard, auditable impact from AI investments: faster closes that show up in working capital, cleaner forecasts that improve guidance accuracy, and measurable savings that hit the bottom line. Anything less (think productivity 'estimates' or vanity metrics) won't make the cut. In 2026, AI must pay for itself, just like any other capital investment," he said.
Divergence
As science fiction author William Gibson noted, the future is already here — it's just not very evenly distributed. While many firms are leaning into automation to address capacity issues, there's still a material proportion of firms that are not, whether this is due to personal preference, budgetary constraints, or the nature of their specific workflows. Regardless, several industry experts have observed that this is creating a divergence in the profession that will likely widen into this coming year.
"AI is accelerating a divergence in firm performance. Firms that have embraced automation are handling more engagements with leaner teams and delivering faster, more modern client experiences. Meanwhile, firms slow to adopt are struggling to keep up with rising complexity and shrinking talent pools. By 2026, we'll see a clearer split between firms that scale strategically with technology and those that are limited," said Jin Chang, CEO of audit solutions platform Fieldguide.
However Schellman's Desai felt it was less a divergence between firms and more between types of work. In her mind, there is no reason firms have to pick one or the other; they're perfectly capable of doing both. "I think we are going to see the profession split into two lanes in a good way," she said. "Lane No. 1 will be highly automated, faster, cheaper, and focused on execution. Lane No. 2, what we are typically used to, advisory and assurance rooted in trust, risk and interpretation. CPAs who thrive will be the ones who lean into the second lane and use the first lane as leverage — truly have a copilot!"
"The value of the CPA goes up, not down, because when AI handles the routine, what's left is the part clients need humans for. Judgment, ethics, and confidence in the numbers," she added.
People still matter
While all these things indicate the profession is rapidly advancing its digital transformation, 2026 will not be the year accountants all lose their jobs. While there are many things AI is capable of doing already, and many more it will be capable of by the end of this year, there will remain a great many things that are the sole domain of human brains.
Jeff Seibert, CEO of AI-driven accounting platform Digits, said many of them fall in the domains of review, judgment and client communication. Like virtually all people we heard from on this topic, these areas, he was confident, won't be automated anytime soon — or at least not this year.
"Even with near-fully automated closes, accountants still need to evaluate edge cases, confirm intent behind unusual transactions, interpret anomalies, and contextualize insights for clients," he explained. "These tasks require professional skepticism, ethical reasoning, and nuanced understanding of the business — areas where generative models and agents can support but not replace. Clients don't want an AI to explain their financial results or guide strategic decisions; they want a human who knows their business, their goals, and the tradeoffs involved. As automation takes over the tedium, these human-centric skills actually become more central to the value accountants deliver."
