All writing
4 min read

What Recent CPA News Says About AI Workflows, Controls, and Automation for Professional Services Firms

Recent accounting-industry news points to a familiar operational theme: firms need tighter controls, better workflow design, and more resilient automation. For law and accounting leaders, that is exactly where custom AI and agentic workflows can reduce risk without adding noise.

AI workflowsautomationrisk controlsaccounting firmslaw firmsprofessional servicesprofessional services AI workflowsaccounting firm automation

Three recent stories offer a useful reminder for firm owners: process gaps, staffing strain, and complex compliance work still create the biggest operational risks. Custom AI and automation are most valuable when they tighten those weak points, not when they simply add another tool.

Operational risk is still a workflow problem

A former nonprofit accountant was accused of embezzling more than $54,000 after a shortage showed up in an annual review of financial records. The organization's executive director reported the issue, and investigators said the case involved computer crime and identity theft charges. For firm leaders, the lesson is straightforward: weaknesses often surface only after money has already moved.

That is why custom AI should be paired with workflow controls, review steps, and exception handling. In a law or accounting firm, the goal is not to let AI make the final call. The goal is to flag unusual activity earlier, route it to the right person, and preserve a clear record of what happened next.

Paper-based work and manual exceptions slow firms down

The GAO review of the IRS filing season showed how staffing and systems issues can create delays when work depends on paper. Individual paper returns took about 30 calendar days to process, compared with 16 the year before, and IRS systems were not available for tax-year 2025 individual paper returns during the first six weeks of 2026.

Professional-services firms face their own version of this problem whenever documents, approvals, and follow-up tasks sit in inboxes or spreadsheets. Agentic workflows can help by triaging incoming items, creating status updates, and moving work to the next step based on defined rules. The value comes from making the process visible and consistent, especially when the work is repetitive but still important.

Compliance complexity is a strong use case for custom AI

The AICPA's September 2026 update highlights ongoing requests for clearer guidance on CAMT issues, including purchase accounting, push-down accounting, section 174A research and experimental expenditures, intangible drilling costs, and CFC income double counting. That is a reminder that tax and advisory teams spend a lot of time translating dense rules into practical client actions.

Custom AI can support that work by organizing source material, summarizing issue lists, and preparing first-pass internal memos for review. For law firms and accounting firms alike, the best use of AI is often not drafting the final advice. It is helping experts move faster from raw guidance to a supervised, client-ready decision path.

Work design matters as much as software design

BILL's move to a remote-first workplace is another sign that firms and platforms are rethinking how work gets done. The company framed the shift around spending less time on commuting and more time supporting clients. That same logic applies inside professional-services firms, where process design can either free up experts or bury them in coordination work.

If a firm is investing in AI, the best starting point is usually a narrow workflow with clear ownership: intake, document classification, exception review, billing support, or compliance tracking. Once the process is defined, custom AI can be evaluated against real outcomes such as turnaround time, error reduction, and staff time saved.

Operator takeaways
  • Use AI to surface exceptions and route work, not to bypass controls.
  • Start with narrow workflows that already create friction: intake, review, billing, and compliance tracking.
  • Treat paper-heavy or inbox-heavy processes as candidates for automation before layering on more tools.
  • For complex tax and advisory topics, use AI to organize and summarize, then keep human review in the loop.
Related Brightline reading

Sources watched

Have a workflow that sounds like this one?

Every engagement starts with a 30-minute conversation. No pitch. No proposal until we understand your problem. If we can't help, we'll tell you.

Get in Touch