The latest news from legal tech and accounting makes one thing clear: firms are moving past curiosity and into operational questions. Leaders now need to decide where AI belongs in the core workflow, how to secure the data around it, and how to prove that it is actually improving the way the firm works.
AI adoption is shifting from optional tools to core workflow
Coverage of Legal Innovators UK highlights a recurring theme in legal innovation: the most meaningful measure of AI adoption is whether it becomes part of the core workflow, not an optional add-on. That is a useful benchmark for any professional-services firm deciding where custom AI belongs.
For law and accounting firms, the practical question is no longer whether a team can test an AI tool. It is whether the firm can redesign a repeatable workflow around it so the process is faster, more consistent, and easier to manage at scale.
Workflow design matters more than tool demos
The legal tech news also points to growing interest in AI-native workflows and real adoption inside firms. That suggests a shift away from isolated prompts and toward structured systems that connect people, process, and technology.
For firm owners, that means starting with one workflow and asking where automation, AI assistance, or agentic steps can remove friction. Intake, research support, matter setup, client communication, and internal routing are all examples of places where custom AI can fit into a controlled process.
Security and trust are now part of the AI conversation
The accounting news makes clear that technology adoption cannot be separated from security. One article notes that firms may feel confident in their protections, while still having gaps in device security, remote wipe capability, and phishing readiness.
At the same time, the AICPA is proposing updates to SOC 2 Trust Services Criteria, with the stated goal of keeping the framework relevant as artificial intelligence, automation, and other emerging technologies change third-party risk. For firms building custom AI workflows, that is a reminder that governance, vendor due diligence, and access control need to be designed in from the start.
What firm leaders should do next
The common thread across these updates is simple: the firms that benefit most from AI will be the ones that treat it as an operating model decision, not a software purchase. That means choosing use cases with clear business value, measuring whether the workflow is actually being used, and tightening controls around data and devices.
Custom AI can help only if it is tied to the way the firm already works. Firms that connect adoption, security, and workflow design will be better positioned than firms that leave AI as a side project.
- Start with one high-volume workflow and redesign it around a clear business outcome.
- Treat security, device controls, and vendor review as part of AI implementation, not a separate project.
- Measure whether AI is embedded in the core workflow, not just being tested by a few users.
Sources watched
- Legal Innovators is Next Week! Clio, Harvey + (Artificial Lawyer)
- What Real AI Adoption Looks Like (Artificial Lawyer)
- 3 Cybersecurity Priorities Before Tax Season (CPA Practice Advisor AI)
- AICPA Proposes Update of SOC 2 Trust Services Criteria to Enhance the Relevance, Usability and Value of CPAs' Reports (CPA Practice Advisor AI)
