A recent report from CPA.com says that semi-autonomous AI bots are already completing bookkeeping workflows start to finish, fully automating the entire process, and tax compliance is not far behind.
This was one of the findings of CPA.com’s 2025 AI in Accounting report, which looked at trends and patterns in the profession’s ongoing relationship with the technology. It said that, at this point, all bookkeeping can theoretically be completely automated through AI agents, which the report confidently said can now complete workflows end to end: bookkeeping agents can categorize transactions, flag anomalies, generate monthly reports and even draft client messages.
What’s more, the report said that simple tax compliance tasks will likely be next, as new solutions can, theoretically, reduce preparation time from hours to minutes—extracting, analyzing and completing returns with high accuracy, requiring minimal human intervention except for final review. Some firms, said the report, are reporting over 80% automation of individual return preparation.
Елена Бутусова – stock.adobe.com
This development has had several effects, such as an increasing shift in focus from tax compliance to tax strategy for firms. With machines taking over more and more tax prep and return processing, humans increasingly are concentrating on broader strategic consulting for tax clients.
“As automation is able to absorb increasingly more prep and review tasks, CPAs are reallocating time toward strategic tax planning, scenario modeling and client coaching,” said the report. “Firms are upskilling staff to interpret AI output and provide deeper guidance.”
The rising automation of tax return prep has shifted hiring priorities. The report noted that “AI fluency” is now commonly required for even entry-level positions, and that firms are developing “AIready associate” programs that combine technical accounting training with AI tools mastery. They’re also partnering with schools to create specialized pre-employment certifications that validate both domain knowledge and technological competency.
Once they are hired, early career professionals are focusing less on the execution of routine tasks as their predecessors were. Whereas previously a firm might have employed a small army of entry-level associates to fill out 1099s assembly line style, today’s early career professionals focus more on developing “AI oversight” capabilities. Similarly, they’re increasingly being evaluated not on task completion, but on value-added analysis, client communication and effective AI collaboration.
This, in turn, will be part of a broader anticipated shift in a firm’s strategic priorities. The focus today on automating tasks, the report said, will lead to comprehensive systems that coordinate across the entire accounting workflow, with the emphasis being not on task execution (that’s what the machine is for), but “ecosystem orchestration.” The idea is that the solutions within the firm’s ecosystem will manage the interplay between client data, regulatory requirements and team capabilities, optimizing resource allocation and process design in real time.
Under such conditions, annual, quarterly and monthly cadences the profession is accustomed to will give way to a world of continuous operations that adjust instantly to new information. Workflows will reconfigure themselves based on changing priorities, emerging risks and resource availability, ideally creating responsive practices that scale efficiency without sacrificing quality.
Audit and advisory a little slower
While bookkeeping tasks can now be completely automated, and tax compliance not far behind, the report said progress was a little slower on automating audit and risk analysis tasks, describing its status as “slow but strategic adoption.” New tools are making inroads, helping firms focus on higher conceptual matters that require their professional judgment, but due to regulatory and liability complexity, innovation in this area is more methodical.
Meanwhile, AI in advisory is considered “the next frontier.” The report noted the emergence of AI-driven forecasting, budgeting and KPI modeling tools, and firms are blending human intuition with AI-generated what-if scenarios to deepen value-based client conversations. The blending of machine insight with human expertise will likely continue, with the binary choice between automation and human judgment eventually dissolving into fluid partnerships where AI handles routine analysis while elevating professionals’ capacity for strategic thinking.
Our recent story about AI in advisory, though, finds that not all advisory services are equal when it comes to the potential for AI disruption. For certain, more transactional type advisory engagements, like simple FP&A, AI is already eating away at their foundations.
ROI still elusive
The report said that, as of the first quarter of 2025, it is still too early for most firms to quantify the full return on investment from their AI initiatives. While firms are reporting clear productivity gains, the specific financial impact is still developing. Regardless of whether or not the gains can be quantified in revenue terms, the report said firms that have embraced AI are beginning to see meaningful operational gains, including up to a 70% reduction in time spent on manual tasks, five times faster review cycles for tax prep and audits, and a two to threefold increase in client capacity without additional headcount.
