Why ‘enterprise knowledge’ is company’s best asset in the AI era

Every organization now has access to the same AI tools: Large language models, automation platforms, and productivity stacks.

This means that the tools have already stopped being the differentiator.

Organizations’ unique assets have now become the domain expertise accumulated over decades, the judgment experienced workforce carries and the institutional knowledge embedded in how teams make decisions. That is what Raconteur and PA Consulting's latest research calls "enterprise knowledge," and it is becoming the real battleground for competitive advantage in an AI-dense market.

The challenge, however, is that while organizations are deploying AI rapidly, they are simultaneously letting enterprise knowledge leak? Without a deliberate human capability strategy, the two trends compound each other.

The knowledge loss problem is already here

Lepaya's Evolution of Skills 2030 research reveals the extent of the risk. 82% of employees believe workforce demands are developing faster than their skills. And with AI accelerating the pace of change, the shelf life of individual expertise is shrinking.

Lepaya's Future of Workforce Resilience report, based on analysis of 2,200 job openings across 110 high-growth European firms, shows the structural consequence playing out in real hiring data. Entry-level roles have dropped below 20% of all open positions and 75% of roles are now mid- or senior-level. As automation replaces entry-level work, the pipeline through which institutional knowledge is built and passed down is thinning. Organizations are not just struggling to capture existing expertise. They are hiring less of the talent that would absorb it next.

The question this creates for L&D is not only how to develop skills people do not yet have, but also how to capture, protect, and continuously activate the expertise that already exists before it walks out the door.

What enterprise knowledge actually means for L&D

PA Consulting's Alwin Magimay frames the problem precisely: "Many organisations have bought the tools but skipped the thinking." The risk is AI deployments that produce pilots and proofs of concept, but none of the scalable, lasting human knowledge and skills that organizations need.

For L&D professionals, this reframes their strategic priorities. How does L&D develop not only build future=proof skills, but also ensure that the human expertise is transferable, applicable, and durable?

That is a key capability and training design challenge. It requires L&D to think about:

• How is judgment transferred, not just static knowledge?
• Where is institutional expertise concentrated, and what is the succession plan if those people leave?
• How does AI amplify existing expertise rather than replace the thinking that produced it?

Raconteur's research notes that competitive advantage now comes "not from the technology itself but from the strategies you use to gain and preserve enterprise knowledge." For L&D, that is both a strategic prompt and a mandate.

The perpetual beta problem

One of the more useful frameworks from Raconteur's piece is what PA Consulting calls the "perpetual beta" mindset. Organizations can no longer operate on industrial-era planning cycles where a capability is built, deployed, and considered done.

Derreck van Gelderen, PA Consulting's global head of AI strategy, describes the shift: 

"You're not just teaching people to use an AI tool; you're teaching them to fundamentally rethink how they do their job with AI in it. It's the difference between 'deliver it and move on' and continually evolving as the data, environment and business changes."

This maps directly onto what Lepaya's State of Skills 2026 identifies as the dominant organizational shift towards continuous capability loops. The organizations pulling ahead are building systems that continuously identify skill priorities, design learning tied to performance outcomes, and track whether behavior has actually changed on the job.

Lepaya's own data reinforces what is at stake. Across Europe, more than 85% of learners who set intentional learning goals successfully applied their newly acquired skills to boost productivity and achieve measurable outcomes. 

That is the human version of perpetual beta: continuous, goal-directed capability development that evolves alongside the business.

The human skills that protect enterprise knowledge

Raconteur describes the required skills shift clearly: from basic analysis to refined judgment; from linear execution to ongoing orchestration; from individual output to working alongside AI agents to identify, prioritize, and protect enterprise value.

Employers at high-growth firms actively prioritizing professionals who combine AI expertise with capabilities machines cannot replicate: ethical judgment, analytical storytelling, and stakeholder management. That is the talent signal. The L&D investment needs to follow it.

As Marlene De Koning, Director of Workforce Transformation at PwC, puts it:

"To lead a people-centered AI future, HR and L&D must go beyond teaching tools and focus on building talent — cultivating the uniquely human skills AI can't replicate and creating cultures where learning is continuous, experimentation is safe, and transformation is something people shape, not endure."

The governance gap is a human capability gap

Raconteur's article gives significant space to data governance: who owns enterprise data, how standards are enforced, how proprietary knowledge is protected from being inadvertently transferred to AI platforms. These are typically framed as technical or legal problems.

They are also human capability problems.

The organizations that manage this well are not just those with better IT architecture. They are the ones where leaders have the judgment to know which AI use cases generate value and which expose proprietary knowledge unnecessarily. Where middle managers can evaluate a use case for both its ROI and its risk. Where employees have enough organizational context to make sensible decisions about what to share and what to protect.

Naomi de Wolf-Melching, Strategic L&D and Talent Consultant at VodafoneZiggo, describes the infrastructure that makes this possible, in Lepaya's Future of Workforce Resilience report: "With the AI Literacy Framework, based on OECD principles, we identify key AI competencies, analyze skill gaps, and guide HR initiatives like training, reskilling, and cultural adoption. This approach ensures AI literacy is embedded structurally across the organization."

Structural embedding, not one-off awareness. That is the difference between organizations that manage enterprise knowledge effectively and those that run good pilots.

Where this is going

The organisations that capitalize on AI will protect and develop their unique human capital. 

Enterprise knowledge - the judgment, expertise, and institutional memory built over years - is essential. This knowledge, however, is threatened by hiring freezes and shrinking entry-level pipelines. 

For HR and L&D, this is the defining challenge of the next few years: Not whether to adopt AI, but whether the human infrastructure around it is strong enough to make it work and to outlast the next wave of disruption after it.

That means treating capability development as continuous and designing for judgment transfer. Sanne Cornelissen, Founder of The Shortcut, puts it simply in Lepaya's Future of Workforce Resilience report: "AI will only be as powerful as the habits behind it. The real shift is not in tools, but in how people learn to think with them."