Past the Buzzwords!

There's a lot of AI in L&D conversations right now. Not so much of it in L&D platforms. Most "AI-powered" learning tools I see are a chatbot in a different hat — bolted on, sitting beside the LMS rather than woven into the learning experience. So when we talk about an AI-Native LMS, the question I'd encourage every HR and L&D leader in the region to ask is a simple one: what does this actually do for my learners and my team on a Tuesday morning?

Here's how I'd answer that.

The pressures L&D is sitting under

Three things have changed quickly, and they're putting real pressure on learning teams across the GCC.

The first is expectation. Month-old PDFs and static e-learning modules don't cut it anymore. Learners want answers in the moment, in the flow of work, in the language they actually speak.

The second is languages. Across the UAE in particular, a single course often needs to land in Arabic and English at minimum, and frequently in five, ten, sixteen languages. Doing that manually is no longer realistic.

The third is ROI. Boards want evidence that learning is moving the dial — completion rates aren't enough, and qualitative feedback that lives in a spreadsheet helps no one.

AI doesn't make these pressures go away. But the right kind of AI, embedded in the right places, makes them workable.

What "AI-Native" should mean

When I say AI-Native, I mean four capabilities that are designed for learning and built into the platform — not a single chatbot wearing four different hats. Each one solves a specific problem L&D teams have been wrestling with for years.

1. A 24-7 tutor that knows your content

Imagine your onboarding learner at 9pm on a Sunday, halfway through a module on your code of conduct, with a question. Today, they either guess, wait, or quit.

A learning-focused AI tutor — grounded in your actual training materials, your policies, your manuals — answers that question in their language, in context. The bit that often gets missed: every question is captured anonymously on the back end. That's gold for L&D. You stop guessing where learners struggle and start seeing it. "People keep asking how to apply for leave" tells you something specific to fix. So does "they're confused between Planner and Project."

2. Practice scenarios that feel real

Knowledge transfer is one thing. Application is another. Studies are pretty clear that the closer practice gets to real life, the better the learning sticks.

AI scenario builders let you set up realistic conversations — an unhappy customer, a difficult conversation with a direct report, a compliance edge case — and let learners practice as many times as they like. The system tracks the conversation, scores them against the goals you set, shows how the other person's mood shifted, and gives detailed feedback on strengths and gaps. An instructor can only run a roleplay once or twice in a classroom. AI can run it hundreds of times, with consistent feedback.

This is where I see compliance, sales and customer service teams getting the most lift. Particularly in regulated sectors.

3. Reflective feedback at scale

End-of-course tick-box forms tell you almost nothing. The qualitative gold is in conversation — "what could be more helpful?", "what would you change?", "what's missing?" — but no L&D team has the bandwidth to interview every learner.

An AI feedback tool does. It has the conversation in whatever language the learner prefers, then consolidates findings across thousands of responses. We're rolling this out for one client across 16 languages. The team gets one consolidated set of insights, not 16 spreadsheets to reconcile.

Done well, this turns feedback from a backwards-looking compliance task into something genuinely useful for course improvement.

4. Recommendations that keep learners engaged

Engagement is the perennial e-learning problem. Once a learner is on your platform, what do they do next?

A Netflix-style recommender solves this in three layers: continue what you started, what's trending across the organisation, and a personalised pick based on your role or department. It's a small thing, but if your catalogue runs to hundreds or thousands of modules, it's the difference between a learner clicking away and a learner enrolling in a second course.

The Arabic-English reality

I'll say this plainly because it matters: a platform that supports Arabic as a translation afterthought is not fit for purpose in this region. Native Arabic — across the learner experience, the AI tutor, the scenarios, the feedback conversations — has to be assumed, not asked for. If you're evaluating an LMS and Arabic is a "we can add that later" conversation, that's a flag.

What to ask when you're evaluating

A few questions I'd put to any vendor claiming to be AI-Native:

What model are you using, and can I bring my own? Enterprises increasingly have their own LLM subscriptions and data residency requirements. The platform should be model-agnostic.

Where does the data sit? Regional hosting matters for many of our clients, especially government.

What does the AI actually learn from? If it's not grounded in your content, it's just a generic chatbot.

How do you handle red-teaming and quality assurance of AI responses? You need visibility into what the AI is telling your learners.

Is the AI integrated, or alongside? An AI tab is not the same as AI woven into the tutor, the scenarios, the feedback and the recommendations.

Where this leaves L&D leaders

The interesting question isn't "do we need AI in our LMS?" — that ship has sailed. The better question is "what should AI actually do for my learners?"

Answer that first. Then look for a platform that solves those problems specifically, in the languages your people speak, with the security and hosting your organisation requires. The hype will keep coming. The capability you need is more boring and more useful than the hype suggests — and it's available now.

Damian Hehir is Co-founder of Capytech, an L&D solutions company headquartered in Dubai. Capytech holds the Dubai AI Seal and works with public and private sector clients across the GCC on e-learning, AI-native LMS deployments and bilingual content development.

Damian’s profile

linkedin.com/in/damianhehir