The content is stored in your LMS. People in your group need something else.
Walk into any enterprise L&D function and you will find a course catalogue at the centre of the operation. It might be modern, well-designed, and full of high-quality content. But its fundamental job has not changed in twenty years. It organises content. It assigns content. It tracks who consumed the content.
The model worked when learning was something employees did in scheduled blocks, mostly to satisfy compliance requirements or complete onboarding. It does not work for the way modern organisations actually need to develop talent.
Skills are shifting too quickly. Roles are too fluid. The half-life of a typical professional skill keeps shrinking, and the World Economic Forum’s Future of Jobs Report 2025 estimates that 39% of existing skills will be outdated or transformed by 2030. Course catalogues, by their nature, lag behind. By the time a course is built, reviewed, and added to the catalogue, the skill it teaches has often already moved on.
Better catalogs won’t fix this problem for the next generation of corporate learning. Instead, AI agents will fix it.
What an AI Learning Agent Actually Does
An AI learning agent is not a smarter LMS. It is a different category of system entirely.
Where a course catalogue waits for learners to find the right content, an AI agent identifies what each learner needs in the context of their role, performance, and career trajectory. It assembles a development experience from multiple sources rather than serving up a single course. It adapts in real time as the learner progresses, the role evolves, or the business priorities shift.
The agent is not an LMS with a recommendation engine attached to it. The orchestration layer decides what learning happens, when it happens, and how it gets delivered. The course catalog is just one of many things that can be used as input, along with tasks at work, peer coaching, micro-learning, AI-generated practice exercises, and outside resources.
For learners, the experience changes from “go find the course you need” to “the right development is already in front of you.” For L&D teams, the work changes from managing content libraries to managing learning ecosystems.
Why the Shift Is Structural
There is a temptation to view AI agents as the next feature in the LMS evolution. They are not. Adding an AI agent to a course catalogue is like adding a search engine to a filing cabinet. The underlying architecture still constrains what the system can do.
Course catalogues are organised around content. They assume learning is a discrete event with a defined start and end, and that the L&D team’s job is to make sure the right content is available when it is needed.
AI agents are organised around outcomes. They assume learning is continuous, contextual, and embedded in the flow of work, and that the system’s job is to produce capability, not to deliver courses.
These are not compatible philosophies. An organisation can run both in parallel during a transition, but the long-term direction is clear. Industry analysts including Josh Bersin and Gartner have flagged the same shift. The fastest-growing categories in enterprise learning technology are not LMS platforms. They are skills intelligence systems, AI-powered talent development tools, and agent-based learning environments.
What Changes for L&D Teams
The shift to agent-based learning changes the role of the L&D function in three concrete ways.
The first change is in what the team builds. Less time goes into producing courses and more goes into curating learning ecosystems, defining outcomes, and ensuring that the agent has access to the right data and content sources to make good decisions.
The second change is in how the team measures success. Course completion rates become a low-priority metric. The questions that matter become whether capability is improving, whether the agent is recommending the right development at the right time, and whether business outcomes are moving as a result.
The third change is in how the team partners with the rest of the business. Agent-based learning depends on integration with HR systems, performance data, and workforce planning tools. The L&D function cannot operate in isolation. It becomes a strategic partner to HR, IT, and business unit leadership.
What Changes for Learners
The learner experience also changes in ways that matter.
In a course catalogue model, the learner is responsible for figuring out what they need, finding the right content, and motivating themselves to complete it. Most do not. According to LinkedIn’s 2025 Workplace Learning Report, the average employee completes only a small fraction of the learning content available to them.
In an agent-based model, the learner is supported by a system that already understands their role, their goals, and their gaps. Development arrives in context. It is shorter, more relevant, and more likely to translate into actual capability change. Engagement increases not because the content is more entertaining, but because it is more useful.
The Catalogue Is Not Going Away. It Is Just Not the Centre Anymore.
Course catalogues will continue to play a role. Compliance training, certifications, and structured curricula still benefit from organised content libraries. But the centre of gravity in corporate learning is shifting away from the catalogue and toward the agent.
The organisations that recognise this early will build the infrastructure now. The ones that wait will spend the next few years explaining to leadership why their learning function cannot deliver the outcomes the business expects.
If your organisation is ready to explore what agent-based learning looks like in practice, we can help.
Talk to our team at zillearn.com/contact-us/
Sources: World Economic Forum. “Future of Jobs Report 2025.” https://www.weforum.org/publications/the-future-of-jobs-report-2025/ LinkedIn. “2025 Workplace Learning Report.” https://learning.linkedin.com/resources/workplace-learning-report Josh Bersin. “HR Technology 2025: The Market Reinvents Itself.” https://joshbersin.com/hr-technology-market/