For twenty years, companies have been chasing the dream of becoming “skills-based organizations.” They built taxonomies and asked employees to self-report what they could do. They mapped jobs to skills, skills to roles, roles to learning paths. And still, most leaders I talk to admit the same thing: the data is stale almost as soon as it is collected.
ServiceNow appears to be asking a more useful question. What if the problem isn’t that companies need a better skills framework? What if the problem is that the entire idea of a static skills framework is obsolete?
In a recent conversation with ServiceNow’s Vice President of Workforce Skills and Talent Readiness Josh Newman, I heard a term that perked my interest: Talent Signature. The idea is simple but impactful. Instead of relying on a frozen snapshot of what someone says they can do, ServiceNow is working toward a dynamic profile of what a person has actually demonstrated through real work, real behavior, and real outcomes. It is less like a résumé and more like a living fingerprint of capability. And I believe this is an early signal of how AI is beginning to change the way companies understand work itself.
"What someone puts on a form tells you who they were when they filled it out," Newman told me, “Talent Signature tells you what they're capable of and where they're ready to land next. That changes every workforce decision that follows."
For years, “skills-based” talent strategy has been treated as the enlightened alternative to job-based workforce planning. In theory, that is right. Job descriptions are too rigid and titles are often misleading and neither tends to truly capture what people actually do.
But the execution of a skills-based framework has often fallen apart. The conventional model depends on getting the skills data right. This has relied on employees self-assessing, managers validating, and HR defining the taxonomy. The system tries to keep up with the business. And by the time the data is clean, the work has changed.
The problem is that this model treats skills as a database problem: something to catalogue and maintain. When they are really an intelligence problem: something to continuously infer from the work itself. The distinction matters. A database tells you what someone was tagged with at a moment in time. An intelligence layer helps you understand what someone can actually do based on evidence. That is the leap ServiceNow is trying to make.
The “Talent Signature” concept starts with a different premise: people reveal their capabilities through the work they actually perform. That means capability isn’t inferred only from a course completed, a credential earned, or a manager’s once-a-year assessment. It is also informed by a broader set of signals: the outcomes a person has delivered, the problems they have solved, the kinds of work they are pulled into, the capabilities they demonstrate in context, and the gaps between what the system thinks they can do and what leaders see them doing in reality. That is the opportunity. AI can help close the gap between what the formal system knows and what the organization is actually learning through work.
ServiceNow's Talent Signature works across three layers. The first captures the basics: who someone is, their role, and their core profile. The second is where skills, credentials, and learning history live as constantly evolving signals rather than a fixed record, alongside collaboration patterns like who someone regularly works with. The third is where AI pulls everything together into insights like where someone's strengths are trending and what they should do next. Together, these layers answer three questions for every individual: who they are, what they know, and what they need next.
The system is already live, primarily within ServiceNow University, which has grown to nearly 2 million registered learners. Of the hundreds of thousands of individuals who hold credentials on the platform, nearly 95% have earned AI credentials specifically, a sign of just how much capability-building is converging around AI. Talent Signature is the intelligence layer underpinning those outcomes and the foundation for expanding personalization across all ServiceNow experiences as the system matures.
The bigger move here is not simply creating better employee profiles. It is connecting three things companies have historically managed separately: What people can do. What work needs to get done. What outcomes the work produces.
Most HR systems were never designed to connect those dots in real time. Learning data lived in one place, performance data in another, and workforce planning in another. Project work somewhere else. And business outcomes were often completely disconnected from the talent system. That separation is no longer sustainable.
In an AI-first organization, leaders need to know where capability exists, where it is emerging, where it is misread, and where it needs to be built fast. They need to know not just who has a skill, but who has demonstrated the ability to apply that skill in a business context.
ServiceNow’s approach points toward a different kind of workforce architecture: a system where work continuously updates the view of talent, and that view then feeds back into decisions about learning, deployment, role design, and workforce readiness. That is a very different operating model than the annual talent review. It means workforce intelligence becomes a living system.
For years, HR leaders have been told they need to become more strategic. HR business partners, in particular, have been promised a future where they spend less time on manual work and more time advising the business. But that future has been hard to realize because the underlying systems did not provide the intelligence required. HRBPs were often stuck stitching together reports, chasing data, and translating fragmented inputs into something useful.
Agentic AI has the potential to change that for HR. Not because AI magically makes HR strategic, but because it can remove the manual drag and surface better insight. In the ServiceNow example, talent readiness data does not sit off to the side. It becomes fuel for broader workforce decisions. It helps leaders understand whether people are ready for new work, where learning should be targeted, and how roles themselves may need to evolve.
That is the new job of HR: not simply managing people processes, but helping the enterprise redesign the relationship between people, AI, work, and outcomes.
One of the most important parts of the ServiceNow conversation was what they are not doing. They are not pretending the system has perfect knowledge. In fact, they are doing the opposite. When the data does not match what leaders observe, that gap becomes valuable. It tells the organization where the inference engine needs to improve.
Picture a project manager whose Talent Signature shows no AI credentials. But their team lead has watched them build and deploy an AI workflow from scratch over the past quarter. The answer is not to blindly trust the system. The answer is to investigate. What is the system missing? What evidence should be added? Assessments are a critical part of the answer, providing structured, formal evidence of demonstrated capability that the system can learn from. And leaders who can see what the data cannot are a functional part of the design. Human judgment, the human advantage, is how the system gets smarter over time.
There are three practices from the ServiceNow story that every CHRO and CEO should pay attention to.
First, stop treating skills as static inventory. Skills data matters, but it is only useful if it is continually refreshed by evidence from real work. The goal should not be a perfect taxonomy. The goal should be a better intelligence loop.
Second, look for roles that are emerging before they are formally named. AI transformation will create new patterns of work before HR has job descriptions for them. Leaders need systems that can detect those patterns early.
Third, connect capability to outcomes. Workforce planning should not ask “Who has what skill?” It should verify “What capabilities are producing the outcomes we need, where do they exist, and how do we build or deploy them faster?”
That is where the real value is.
One of the biggest constraints in AI transformation is organizational capability. Companies are moving faster than their talent systems can understand. Work is changing and teams are evolving as agents join as team mates. The old machinery of workforce planning cannot keep up.
ServiceNow’s work is important because it points to what comes next. The future is a living intelligence layer that understands people through the work they do, learns from outcomes, and continuously feeds that intelligence back into the business. That is the real promise of AI in HR, finally helping organizations see what their people are truly capable of and redesigning work around that truth.
Originally published at Forbes