Professional services firms risk losing clients, talent, and significant revenue by failing to operationalise AI effectively, according to new research from Thomson Reuters. The company’s 2026 Future of Professionals report, based on a global survey of 1,800 professionals across legal, tax, audit, and compliance functions, puts up to US$143 billion in U.S. client revenue under active reconsideration by clients dissatisfied with their providers’ AI delivery.

The findings point to a structural gap between AI adoption and AI execution — one that is now producing measurable financial and talent consequences for firms that have not moved from strategy to implementation.

Shadow AI and Unmanaged Risk

Despite high rates of AI tool usage — 91% of professionals in Asia and Emerging Markets report using AI weekly — organisations are struggling to convert that activity into verified, accountable output. One-third of lawyers, accountants, and compliance professionals across the survey say they are using AI tools their organisations have not approved, a figure that rises to 41% among those who believe their firm is moving too slowly.

The risk profile of that behaviour is significant. Among all respondents, 96% say their AI must safeguard confidential data, 94% require verified authoritative content, and 90% need outputs they can explain and defend. Yet 41% say they lack access to professional-grade tools that meet these standards — creating what the report characterises as invisible, unmanaged risk at the workload level.

Talent and Client Pressure Converging

The execution gap is beginning to affect workforce stability. One in four professionals say they would consider leaving their organisation within two years if AI delivery does not improve; 13% place that timeline within 12 months. Among professionals already using AI tools, nearly one in three say they would decline a new role that did not offer equivalent access. Thomson Reuters notes that nearly half of senior leaders believe meaningful talent pressure is still at least three years away — a significant mismatch in perception.

Client expectations are equally acute. Some 78% of corporate clients now consider AI-enabled quality improvements very important or essential, yet just 6% believe most of their current providers are delivering. Within 12 months, 32% of clients expect to reassess provider relationships, with a third of those putting more than US$1 million in annual work at risk.

“Firms that are operationalising AI are pulling ahead. Those that aren’t are starting to take on real risk, across talent, clients, and financial performance,” said Steve Hasker, President and CEO of Thomson Reuters. “Closing that execution gap is now a business imperative for professional firms.”

The Case for Fiduciary-Grade AI

Thomson Reuters frames the solution around what it terms Fiduciary-Grade AI — defined by authoritative domain-specific content, rigorous privacy and security, subject-matter expertise, verifiable outputs, and access to real-time human support. The positioning reflects a broader argument that general-purpose AI tools are insufficient for professions where outputs carry legal, regulatory, or advisory liability.

In Asia and Emerging Markets, 90.3% of professionals surveyed believe their organisations are falling short of what AI can deliver. The report’s Asia-specific data underscores that the execution gap is not a problem confined to slower-moving Western markets — it is present and accelerating across the region.