Organizational GenAI adoption in professional services nearly doubled in one year, jumping from 22% to 40%. That is the top-line finding from the Thomson Reuters 2026 AI in Professional Services Report, a survey of 1,514 professionals across legal, tax, accounting, and risk functions in 27 countries. But here is the number that should worry every managing partner and CFO: only 18% of those organizations actually measure whether their AI investments are producing returns. Professional services firms are adopting fast and measuring almost nothing.
Thomson Reuters frames 2026 as the “year of AI impact,” following the experimentation phase of 2024 and implementation phase of 2025. The data backs that framing. 82% of current GenAI users now engage with it weekly or more. 87% expect it to become central to their workflow within five years. This is not a niche experiment anymore. It is a mainstream shift happening without the governance structures to match it.
Adoption Has Doubled, But Strategy Has Not
The adoption numbers are striking when you lay them out year-over-year. In 2025, 22% of professional services organizations used GenAI. In 2026, that hit 40%. Only 19% now say they have no plans to adopt at all, down from a much larger holdout group the previous year.
Individual usage runs even higher. Over 50% of professionals use publicly available tools like ChatGPT, Claude, and Gemini, whether or not their employer provides an official solution. Among those who do use GenAI, the frequency is high: 30% use it multiple times per day, another 25% daily, and 26% weekly. This is not “trying it once and forgetting.” This is a tool people reach for constantly.
The Strategy Gap Is the Real Story
But adoption without strategy is just spending. Only 22% of organizations have achieved what Thomson Reuters calls “strategic clarity” on AI. That means 78% are deploying tools without a clear plan for how AI fits into their service delivery, pricing, or talent models.
The consequences are measurable. Organizations with a defined AI strategy are 2x more likely to experience revenue growth and 3.5x more likely to realize critical AI benefits. There is also a disconnect between personal and organizational goals: 65% of professionals who have personal AI adoption targets report that they have no awareness of their organization’s AI strategy. Another 38% whose organizations do have a strategy say they have set no personal adoption goals. The left hand and right hand are not talking.
“2026 marks the strategic phase of AI, one where organisations redefine workflows, reshape value, and build AI directly into the foundation of their business strategy,” said Mike Abbott, Head of the Thomson Reuters Institute. That is aspirational. The data suggests most firms are not there yet.
Where Professional Services Actually Use AI
The use cases differ by function, and the data here is genuinely useful for anyone deciding where to deploy next.
Legal
Legal professionals have clustered around research and review. 80% use AI for legal research, 74% for document review, 73% for document summarization, 59% for drafting briefs and memos, and 49% for contract work. The bias toward research makes sense: it is the use case with the clearest accuracy feedback loop. You can check whether the AI found the right cases. Drafting, where errors are harder to catch, lags behind.
Tool choice tells its own story. 55% of legal users rely on general-purpose tools like ChatGPT, while only 35% use specialized legal AI tools. 38% use enterprise platforms like Microsoft Copilot. The specialized tools should outperform general-purpose ones on legal tasks, but the convenience gap keeps ChatGPT in the lead.
Tax, Accounting, and Risk
Tax and accounting firms show a similar pattern at lower intensity: 69% use AI for tax research, 57% for document summarization, and 55% for document review. The Wolters Kluwer Future Ready Accountant survey found AI adoption in accounting jumped from 9% in 2024 to 41% in 2025, with revenue per employee at tech-optimized firms hitting $250K-$350K versus $150K-$200K at traditional firms.
Risk and fraud teams are the heaviest users in the document intelligence category: 86% use AI for document summarization, 74% for document review, and 71% for risk assessment. These teams deal with high-volume, pattern-recognition workloads that are a natural fit for LLMs.
Agentic AI: 15% In, 53% Watching
The agentic AI numbers are the most forward-looking data point in the report. Only 15% of professional services organizations currently use agentic AI, systems that can plan, execute multi-step workflows, and take action autonomously rather than just generating text when prompted.
But the pipeline is substantial: an additional 53% are planning or considering agentic deployments. In legal specifically, 16% currently use agentic AI, 19% plan to, and 30% are considering it. Only 35% have no plans at all. That means 65% of legal firms have agentic AI somewhere on their roadmap.
77% of professionals surveyed expect agentic AI to become central to their workflow by 2030. That is four years from now, which in technology adoption terms is not distant at all.
Thomson Reuters is putting its own chips on this bet. CoCounsel, their AI assistant built on the Casetext acquisition ($650 million in cash in 2023), has reached 1 million users across 107 countries. Roughly a quarter of Fortune 1000 companies now use it. The next-generation features entering beta include conversational task execution (describe an objective and the system builds a plan, retrieves authority from Westlaw, searches documents, verifies citations, and delivers structured work product) and bulk review of up to 10,000 documents. These are agentic capabilities, not just chat.
The $32 Billion Time Savings Nobody Measures
The projected economic impact is large. Legal professionals expect AI to free up nearly 240 hours per year, up from 200 hours in 2024. That translates to about 5 hours per week and an average value of $19,000 per professional per year. Across the US legal and tax/accounting sectors combined, Thomson Reuters estimates the annual impact at $32 billion.
These are projections, not measurements. And that is the problem.
The ROI Measurement Gap
Only 18% of organizations in the survey actually measure AI ROI. 40% of respondents do not even know whether their organization tracks ROI at all. Among those who do measure it, 77% track cost savings and 64% track employee usage, but fewer than 30% measure client satisfaction or new business generation.
This gap matters because the firms that do measure are seeing differentiated results. Organizations with a formal AI strategy are 3x more likely to achieve positive ROI. 53% of professionals report seeing hard returns in time saved and error reduction. The returns appear to be real, but most firms are flying blind on whether they are capturing them.
The comparison to broader enterprise AI adoption is instructive. The McKinsey State of AI survey found that only 6% of organizations qualify as “AI high performers” generating more than 5% of EBIT from AI. Forrester has warned that enterprises will defer 25% of planned AI spend into 2027 due to ROI concerns. Professional services are not immune to this pattern.
What Clients Actually Expect (And What Firms Fear)
One of the report’s more interesting sections covers the client-firm dynamic around AI. 74% of corporate tax clients believe their firms should be using AI. 54% of corporate legal clients agree. But fewer than 20% mandate AI use through RFPs or formal guidelines. Client demand is strong but informal.
Threat perceptions among professionals have spiked year-over-year. The share viewing AI as a potential vehicle for unauthorized practice of law jumped from 36% in 2025 to 50% in 2026. Those who see AI as a direct threat to jobs went from 15% to 24%. Those expecting fewer lawyers will be needed rose from 10% to 18%.
91% of professionals believe AI should be held to higher accuracy standards than humans, and 41% say AI outputs need 100% accuracy before they can be used without human review. That 100% accuracy bar is, of course, impossible for any system, human or machine. It signals a trust deficit that no amount of capability improvement will satisfy without institutional frameworks around verification and accountability.
As David Wong, Thomson Reuters’ Chief Product Officer, put it: “Speed is important, but trust is even more valuable. No matter how advanced agentic capabilities get, they won’t be adopted if professionals can’t trust them.”
Steve Hasker, Thomson Reuters President and CEO, framed the shift bluntly: “Professionals are not deciding whether to use AI anymore. They are deciding which AI they trust when their reputation and their clients’ data are on the line.”
Frequently Asked Questions
What does the Thomson Reuters 2026 AI report say about adoption rates?
Organizational GenAI adoption in professional services nearly doubled from 22% in 2025 to 40% in 2026, based on a survey of 1,514 professionals across legal, tax, accounting, and risk functions in 27 countries. 82% of current users engage with AI weekly or more, and 87% expect it to become central to their workflow within five years.
How many professional services firms use agentic AI?
Only 15% of professional services organizations currently use agentic AI. However, an additional 53% are planning or considering it, and 77% of professionals expect agentic AI to become central to their workflow by 2030.
What is the projected ROI of AI in professional services?
Legal professionals expect AI to free up nearly 240 hours per year, worth an average of $19,000 per professional. Thomson Reuters estimates the combined annual impact for US legal and tax/accounting sectors at $32 billion. However, only 18% of organizations actually measure AI ROI.
What are the top AI use cases in legal firms?
Legal research leads at 80% adoption, followed by document review at 74%, document summarization at 73%, drafting briefs and memos at 59%, and contract work at 49%. 55% of legal users rely on general-purpose tools like ChatGPT rather than specialized legal AI solutions.
Do clients expect law firms and accounting firms to use AI?
Yes. 74% of corporate tax clients and 54% of corporate legal clients believe their firms should be using AI. However, fewer than 20% mandate AI use through formal RFPs or guidelines, making client demand strong but largely informal.