We are witnessing a fundamental shift in artificial intelligence that will transform how professionals work. The emergence of agentic AI—systems capable of autonomous reasoning, decision-making, and task execution—represents the next evolutionary leap beyond today’s generative AI tools. But as this technology matures, a critical reality is becoming clear: not all companies are positioned to deliver truly reliable, professional-grade agentic AI.
The difference lies in four essential components that agentic AI requires to function effectively in professional environments: advanced reasoning models, comprehensive domain-specific content and expertise, and access to the tools professional use. Companies that possess all four will thrive; those that lack even one will struggle to compete.
The four pillars of professional agentic AI
First, advanced reasoning models form the cognitive foundation. These aren’t simply large language models trained on internet data—they’re sophisticated systems capable of multi-step reasoning, logical inference, and contextual understanding within specific professional domains.
Second, comprehensive, authoritative content and expertise is crucial. Agentic AI systems must access vetted, structured information that professionals can trust. This isn’t just about having large datasets; it’s about having the right data, that’s properly curated and continuously updated.
Third, you also need subject matter experts – lawyers, tax professionals, risk specialists – to provide the human intelligence that trains, validates, and guides these systems. This expertise ensures that agentic AI understands not just what information exists, but how it should be applied in real-world professional contexts. If you think of content as being the book smarts, subject matter expertise is the street smarts guiding how work gets done, resolving problems and nuances in professional work.
Four, professional-grade AI assistants need access to the same tools that professionals use to perform professional work. Without tool access, AI assistants can only provide advice or suggestions—you cannot delegate actual tasks because the AI lack the means to complete work successfully. An example is if you want AI to solve tax problems, it needs to know how to use a calculator.
The scarcity of these four components is already reshaping our industry. We’re seeing major consequences that underscore this reality.
Strategic partnerships and M&A activity
We’re seeing a growing wave of strategic partnerships and acquisitions. Partnerships between AI providers and content owners, like Harvey and LexisNexis, are becoming increasingly common. AI companies are seeking access to authoritative content and domain expertise they cannot develop internally, while established information providers are partnering to enhance their AI capabilities. Few companies are like Thomson Reuters and already have access to both. We’re likely to see more of these alliances as companies recognize they don’t have all four components, and they’re essential.
The talent war intensifies
Perhaps nowhere is the competitive pressure more visible than in the fierce battle for talent. The intensity of this competition has reached new heights. We’ve seen the disputes between Meta and OpenAI, where both companies are aggressively recruiting each other’s top researchers and engineers. But the war for talent has evolved beyond technical roles, and now the battles is heating up for the subject matter experts that provide the judgment and experience that trains, validates and guides these systems. Thomson Reuters has more than 4,500 subject matter experts, including lawyers and accountants, but this is a gap many organizations are now trying to close so they can build agentic tools that work the ways professionals do.
Transforming organization structure and roles
The third major consequence we’re witnessing is perhaps the most transformative: companies are fundamentally reimagining organizational structures and creating entirely new categories of professional roles. As organizations scramble to acquire the agentic AI trifecta, they’re discovering that traditional job descriptions and reporting structures simply don’t work. This shift is creating unprecedented demand for new hybrid roles: prompt engineers who refine human-AI interactions, agent orchestrators who manage complex AI workflows, and human-in-the-loop specialists who handle exceptions and ensure quality.As companies evolve to meet the needs of agentic, they’re rethinking everything from career progression to performance metrics in an age where success depends on human-AI collaboration. Thomson Reuters began this work more than two years ago, and we’re constantly adapting, making sure our teams are aligned to deliver the most advance agentic AI for our customers.
Looking ahead
The agentic AI revolution will accelerate market consolidation and force strategic decisions across our industry. Companies will need to choose: build these capabilities internally, partner with those who have them, or risk being left behind.
The winners will be those who recognize that agentic AI isn’t just about algorithms—it’s about the intersection of advanced technology, authoritative content and expertise, and professional tools. As this technology reshapes how professionals work, the companies that invest in all four will define the future of professional services.
The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.
