Blogs, News & Insights

AI in learning is overpromised. Here's what actually delivers

Written by Philip Huthwaite | 16 April 2026 09:05:27 Z

AI is everywhere in learning right now. Every platform, every vendor, every roadmap seems to have it front and centre.

But as the latest research from Fosway Group makes clear, there’s a growing gap between what’s being promised and what’s actually being delivered.

For learning leaders, that gap matters. The decisions you make today about platforms, partners and strategy will shape how effectively your organisation builds skills tomorrow.

So what does the reality of AI in learning systems look like, and what should we do about it?

 

1. The AI hype cycle is real, and it’s distorting decision-making

Fosway’s analysis highlights something many of us have suspected: the conversation around AI in learning is advancing much faster than real-world capability.

In fact, only around 10% of roadmapped AI capabilities in learning systems are actually live with customers today.

It means that much of what’s being sold to organisations is still conceptual, early-stage or sitting on a roadmap somewhere in the future.

For buyers, the implication is simple: don’t buy the promise, buy the proof. That means real products, available to try and buy right now.

 

 

2. Most innovation is clustered, not transformational (yet)

Another key insight is that AI investment is heavily concentrated in a few ‘hot zones’. Primarily:

  • Learning experience (personalisation, search, recommendations)
  • Learning management (automation, admin efficiency)

These are valuable improvements, but they’re largely incremental enhancements, not wholesale transformation. Once these become commonplace in live learning platforms, the differentiation between these AI features will be relatively minor. AI automation is, broadly, the same across learning platforms, and this isn’t where the exciting stuff is happening.

The more ambitious opportunities around skills intelligence, organisational insight and workforce capability building, are still emerging for the vast majority of vendors, but this is where the real long-term value lies.

 

3. Roadmaps are not reality, and shouldn’t be treated as such

One of the most important takeaways from Fosway’s work is this: roadmaps are not delivery plans. In theory, a vendor could put absolutely anything on a learning technology roadmap, but that doesn’t mean it will ever come to fruition.

A significant proportion of AI capabilities are either:

  • Not yet built
  • Not yet piloted
  • Or not even on near-term roadmaps at all

For learning leaders, this creates risk. If you’re investing in learning systems based on a roadmap, you may find that a year down the line, those promised features ended up on the virtual scrapheap. A roadmap isn’t a legally binding contract, and vendor ambitions and priorities can change – often out of sync with your own.

The smarter approach is to anchor your strategy in what is live, proven and delivering value today, while keeping an eye on what’s next.

 

4. AI is already reshaping learning, but only when applied with intent

Despite the noise, one thing is clear: AI is not a passing trend.

Fosway’s wider research shows that AI is now a core strategic priority for learning teams, fundamentally changing how work is done and what people need to learn.

But impact doesn’t come from simply ‘adding AI’. It comes from applying AI in ways that:

In other words, learning teams shouldn’t be investing in AI just for the sake of investing in AI. AI only matters when it solves real problems, whether that’s filling content gaps, offering on-demand coaching or supporting continuous skills development.

 

5. The winners will be those who execute, not just innovate

This is where the market is starting to split.

On one side, there are vendors with compelling AI narratives. They’re talking the talk, but when it comes down to it, where is the live technology to back up the narrative?

On the other, there are those actually launching, iterating and learning from real-world use.

At 5app, we’ve been very deliberate about which side we want to be on.

We’re not interested in AI as a marketing story. We’re focused on AI as a product reality, which is why we’re constantly delivering updates and improvements to our three AI products: VeeCoach, our AI coach, VeeCreate, our AI authoring tool and Helix, our AI skills intelligence platform.

That means:

  • Building a clear AI roadmap grounded in customer needs
  • Prioritising features that are live, usable and delivering value now
  • Iterating quickly based on real usage and feedback
  • Being transparent about what’s available today vs. what’s coming next

We’re already embedding AI into how organisations:

  • Surface the right learning at the right time
  • Connect learning to skills and performance
  • Reduce noise and cognitive overload
  • Support employees in the flow of work

And crucially, we’re doing this in a way that aligns with what Fosway’s research is calling for: evidence over hype, delivery over promise.

 

So what should learning leaders do next?

If there’s one thing to take from Fosway’s assessment, it’s this:

AI in learning is real, but it’s uneven, immature and often misunderstood.

To navigate that, focus on three principles:

  1. Interrogate the reality – What’s live? Who’s using it? What’s the impact?
  2. Prioritise value over novelty – Does it solve a real problem for your people?
  3. Partner with executors – Choose vendors who are building, not just talking

The organisations that succeed with AI won’t be the ones who bought the boldest vision. They’ll be the ones who implemented practical, working solutions and scaled them with intent.

At 5app, we believe AI should make learning simpler, smarter, and more human, not just flashier or trendier.

And we’re committed to proving that – not just talking about it.