Can AI Help Learning Stick?
According to the 2026 HR & L&D Trends Report, only 24% of organisations report strong application of new learning in the workplace. While businesses continue to invest heavily in training, leadership development, and workforce capability, many still struggle with the same challenge: creating real, tangible behaviour change in learning and development.
This presents an uncomfortable reality for Learning & Development leaders. The issue is rarely the quality of the programme itself. Nor is it a lack of enthusiasm from learners. More often, the challenge lies in what happens afterwards.
An employee attends a leadership programme, participates actively, and leaves with a clear understanding of what effective leadership looks like. They intend to delegate more effectively, coach their team more frequently, or approach difficult conversations with greater confidence. Yet as workloads increase and priorities compete for attention, those intentions are often overtaken by familiar routines and existing habits.
The result is a gap between learning and application.
As organisations face accelerating technological change, increasing pressure on productivity, and a workforce that must continuously adapt to new skills requirements, this gap is becoming one of the most important capability challenges of the AI era.
At Instep, we believe the organisations that thrive over the next decade will not be those that provide the most learning. They will be those that are most effective at helping people apply learning in the real world. The challenge is no longer access to knowledge. The challenge is turning knowledge into capability.
This is where AI agents are beginning to generate significant interest within the L&D community.
Not because they can replace great coaches, mentors, or managers, but because they offer a practical way to reinforce learning long after a programme has ended.
We’ve Solved Access to Learning. We Haven’t Solved Application
For much of the last twenty years, Learning & Development has focused on access.
Organisations have invested heavily in learning management systems, digital platforms, virtual classrooms, on-demand content libraries, and more recently, AI-powered learning tools. The objective has been clear: make learning available to more people, more often, and in more flexible ways.
In many respects, this objective has been achieved.
Today’s workforce has unprecedented access to information. Employees can find guidance, best practice, technical knowledge, and expert opinion within seconds. Increasingly, AI tools are making that process even faster.
However, access to information and capability development are not the same thing.
A first-time manager may understand the principles of effective delegation but continue to struggle with letting go of tasks. A team leader may be able to explain the benefits of coaching conversations while still defaulting to giving instructions under pressure. A senior leader may understand the theory behind strategic thinking but find it difficult to create the time and space required to practise it.
The challenge facing modern L&D is therefore not simply helping people learn. It is helping them embed new behaviours consistently enough for those behaviours to become habits.
That is a far more difficult problem to solve.
Why Learning Transfer Remains So Difficult
Behavioural change does not happen during a workshop.
It happens afterwards.
Most learners do not require additional motivation immediately after a programme. In fact, they are often at their most motivated. The difficulty emerges in the weeks and months that follow, when good intentions collide with the realities of everyday work.
A manager who has committed to improving their coaching skills may find themselves under pressure to deliver results quickly and revert to telling rather than coaching.
An aspiring leader who has learned the importance of active listening may find themselves jumping straight to solutions when time is limited.
Even highly effective programmes can struggle to create lasting impact when learners are left without opportunities to practise, reflect, receive feedback, and remain accountable.
This challenge is often referred to as learning transfer: the process of applying learning in real workplace situations and sustaining the resulting behavioural changes over time.
Historically, organisations have relied on line managers, mentors, and professional coaches to support this process. While highly effective, these interventions are difficult to provide consistently at scale.
This is where AI may offer something genuinely valuable.
What Do We Mean by an AI Agent?
The term “AI agent” is often used broadly, which can make it difficult to understand its practical relevance.
In simple terms, an AI agent is an AI-powered assistant configured to fulfil a specific purpose.
Within Learning & Development, that purpose might be to act as:
- A leadership development coach
- A reflection partner
- An accountability assistant
- A communication skills mentor
- An action-planning guide
- A scenario-based practice partner
Many organisations can begin experimenting with these approaches using familiar tools such as Microsoft Copilot, Claude, or ChatGPT.
The key difference between an AI agent and a traditional chatbot interaction is intent. Rather than answering a single question, an AI agent supports an ongoing development process. It is designed to help learners return repeatedly to their goals, progress, challenges, and behavioural commitments.
The objective is not to provide more information.
The objective is to improve application.
Turning Learning Into Action
One of the most common weaknesses of certain leadership and professional development programmes is the transition from insight to action.
Participants often leave training with pages of notes and a long list of ideas, yet relatively little structure for what comes next.
This is an area where AI can provide immediate value.
Imagine a manager has just completed a leadership programme with a particular focus on delegation. Rather than simply filing away their workbook, they could ask an AI tool:
“I’ve completed a leadership programme and realised I need to delegate more effectively. Help me build a 30-day action plan with weekly goals, measurable outcomes, reflection questions and accountability checkpoints.”
The result is unlikely to be perfect, nor should it be followed blindly. However, it provides a practical starting point that transforms a broad aspiration into a more structured commitment.
Rather than leaving development to chance, the learner has a framework for application.
Encouraging Reflection and Accountability
Reflection is one of the most effective ways to accelerate professional growth, yet it is often the first activity to disappear when workloads increase.
Most professionals recognise the value of taking time to review what is working, what is not, and what they might do differently. The challenge is finding the discipline to maintain that habit consistently.
AI agents can help by introducing regular prompts for reflection.
For example, a learner might ask:
“Act as my leadership coach. Ask me five questions about how I’ve applied my learning this week. Challenge my thinking where appropriate and identify one area for improvement.”
While the AI is not providing genuine coaching in the traditional sense, it is creating the conditions for reflection. Over time, those regular moments of reflection can strengthen self-awareness and increase the likelihood that new behaviours are sustained.
What matters is not the technology itself. What matters is the habit the technology helps create.
Providing Support in the Flow of Work
Another limitation of traditional training is timing.
Learning often takes place far away from the moment it is actually needed.
Someone attends a workshop in January but faces a difficult leadership challenge in March. By that point, key concepts may have been forgotten or become difficult to retrieve.
AI changes that dynamic.
A manager preparing for a challenging performance conversation can seek support immediately before the discussion takes place.
For instance:
“I’m about to have a conversation with a team member who has consistently missed deadlines. Help me structure the discussion, suggest coaching questions, and highlight common mistakes I should avoid.”
The manager still needs to exercise judgement. They still need to understand the context, the individual involved, and the wider organisational environment.
However, they no longer need to rely entirely on memory.
Learning becomes available at the point of need, where it is most likely to influence behaviour.
Creating Safe Opportunities to Practise
One of the most promising applications of AI within workplace learning is its ability to create opportunities for practice.
Many leadership and management skills are difficult to develop because the opportunities to practise them are limited and often carry real consequences.
Difficult conversations are a good example.
Few new managers actively look forward to delivering constructive feedback or addressing performance concerns. Yet these moments are critical to effective leadership.
AI can provide a low-risk environment in which learners can rehearse.
A manager might ask:
“Act as a team member who is frustrated about a recent decision. Allow me to conduct the conversation and then provide feedback on my communication style.”
The interaction will never fully replicate a real conversation, but it does allow learners to build confidence, experiment with different approaches, and identify blind spots before entering a higher-stakes situation.
For many organisations, this represents an entirely new way of supporting skill development between formal learning interventions.
The Limits of AI
As organisations explore these opportunities, it is important not to confuse reinforcement with replacement.
AI can play a valuable role in helping learning stick. It can encourage reflection, support action planning, facilitate practice, and provide reminders at scale.
What it cannot do is replicate the human elements of development.
Trust, empathy, judgement, lived experience, and meaningful challenge remain fundamentally human capabilities. Great coaches bring perspective. Great managers understand context. Great mentors draw on years of experience that cannot simply be replicated through a prompt.
The most effective organisations are therefore unlikely to choose between AI and human development.
Instead, they will use AI to strengthen the spaces between human interactions. They will use it to extend the impact of programmes, reinforce learning, and create greater consistency in development journeys.
In short, they will use technology to support capability building, not replace it.
The Future of L&D Is Application
The conversation around workplace learning is changing.
For many years, the focus was on providing more content, more programmes, and more learning opportunities. Those priorities remain important, but they are no longer enough.
Knowledge is increasingly abundant. Capability is not.
The organisations that gain a competitive advantage in the AI era will be those that become better at helping people translate learning into performance. They will focus less on attendance and more on application. Less on completion and more on behaviour change.
AI agents are not a silver bullet, and they are certainly not a substitute for high-quality leadership development. However, they may help organisations address one of the oldest and most persistent challenges in Learning & Development: ensuring that what is learned is actually used.
Ultimately, the future of L&D will not be defined by how much content organisations can provide.
It will be defined by how effectively they help people develop capability, apply new skills, and perform at a higher level.
AI can help make that possible.
But the outcome that matters remains unmistakably human.
Building Human Capability in the AI Era
As AI makes knowledge more accessible than ever, competitive advantage increasingly comes from applying that knowledge effectively.
That’s why we’re focused on helping organisations build the leadership, communication, critical thinking, and behavioural capabilities that technology alone cannot create.
Ready to explore a more human approach to workforce development in the AI era?
Get in touch here.
