The Capability Paradox
Why AI Makes In-Person Training More Important
AI is often positioned as the answer to one of today’s most frequent challenges in commercial capability: how do you deliver consistent and scalable learning to dispersed teams without getting everyone in a room? An increasingly common response is to utilise the power of AI to build a platform, populate it with content and let people learn on demand. Efficient, scalable and measurable.
Broad reach is not the same as deep impact. In the rush to democratise capability through AI, many organisations are quietly trading lasting impact for basic coverage.
What AI Does Well
To be clear, AI-powered learning has genuine value. It can deliver consistent frameworks to a global team overnight. It can test knowledge retention, remove the barrier of geography entirely and even adapt to an individual’s learning pace. Its ability to respond with follow-up questions and generate examples on demand, combined with its relatively low barrier to entry compared with traditional e-learning, makes it a powerful learning tool, but not one without limitations.
Where It Falls Short
The problem is that commercial capability isn’t primarily a knowledge problem. It’s a judgement problem. The gap between knowing a framework and being able to apply it in a retailer meeting, category review or commercial negotiation isn’t closed by completing a module, reviewing case studies or simply reframing concepts. It’s closed by doing it, sharing experiences, getting it wrong in a safe environment, reacting to ever-changing dynamics and being guided through why.
That process requires human facilitation. It requires a room where someone can challenge an assumption, where a peer’s question reframes your own thinking and where a facilitator can read the dynamic, share real experiences, apply them to specific situations and adjust in real time.
The moments that actually change how people work are rarely the moments of content delivery. They are the moments of application: the case study that mirrors a real challenge, the role play that exposes a gap and the conversation that shifts a perspective.
Fifteen years of delivering capability programmes globally has shown us repeatedly that, while AI can greatly enhance capability programmes, it cannot replicate these critical human interactions.
The Paradox
If we reflect on the classic 10:20:70 learning framework, it becomes easier to articulate the potential limitations of a heavily AI-driven capability programme. The 10% formal content can be delivered efficiently, consistently and at scale, while elements of the 20% coached component can be covered with well-structured AI tools. However, the crucial link to the 70% experiential component, which should include on-the-job consultation, shared experiences and peer consultation, remains untouched.
Here is the counterintuitive conclusion: as AI handles the foundational knowledge layers more efficiently, it raises the bar for what in-person capability needs to deliver. The organisations using AI well aren’t replacing their training programmes. They’re using it to accelerate the baseline so that, when teams do come together, the conversation can start higher and go further. AI does the knowledge transfer. In-person does the capability shift.
That distinction matters more now than it ever has, because the commercial environment is more complex, more competitive and less forgiving of teams that know the theory but can’t apply it in the room.
