Every national AI strategy eventually collides with the same wall: there are not enough people who can build the thing. Compute can be procured in months. Capable engineers take years.
Saudi Arabia’s response was SAMAI, a national AI upskilling initiative that reported training more than one million participants in a single year. Whatever the depth of any individual course, the scale of that number is unusual by international standards.
This article looks at why skills are the real constraint, what mass upskilling can and cannot achieve, and how employers in the Kingdom should respond to a labour market that is changing faster than it is growing.
We Build Your Team’s AI Capability While We Build the System
Elbetron Technologies is a Saudi technology company building production AI for organisations in the Kingdom and the wider GCC. Our work is not demos — it is systems that answer real customers, in Arabic and English, every day.
Elbi, our bilingual AI assistant platform, is the clearest example: retrieval-grounded answers drawn from your own documents, deployable on infrastructure you control, with a voice layer for phone and in-app conversations.
From first workshop to production rollout, we design, build and run the AI systems behind Saudi customer service, operations and internal knowledge.
Skills Are the Real Constraint
Hardware is a purchase order. Models are a licence. Neither is a durable advantage, because your competitor can sign the same contracts. The scarce input is people who can take an AI system from a promising demo to something that survives contact with real users.
That scarcity is global, which makes it expensive. Countries that grow their own talent reduce their exposure to an international bidding war they cannot win on salary alone — and Saudi Arabia has been explicit about wanting capability, not just deployment.
- Engineers who can productionise, monitor and debug models
- Data engineers who make internal knowledge retrievable
- Domain experts who can judge whether an answer is actually correct
- Governance people who understand both the law and the system
What Mass Upskilling Actually Achieves
A one-week course does not create a machine learning engineer, and it is not meant to. What large-scale programmes reliably produce is fluency: a workforce that understands what AI is, where it fails, and how to work alongside it without either fearing it or over-trusting it.
That fluency has compounding effects. When a finance manager knows a model can hallucinate a number, they ask for the source. When an operations lead knows what automation can absorb, they scope projects that succeed. Adoption failures are usually organisational, not technical.
Where the Pyramid Still Narrows
Broad fluency does not solve the top of the pyramid. Deep specialists — Arabic NLP researchers, evaluation experts, people who have run large models in production — remain scarce, and no short course produces them. That layer is built over years and defended with retention, not recruitment.
For employers this means a two-track approach: use national programmes to raise the floor across your organisation, and treat your small number of deep specialists as strategic assets. Losing one of them costs far more than the salary difference that would have kept them.
- Use national programmes to build organisation-wide fluency
- Protect and develop the few deep specialists you have
- Partner for capability you cannot hire fast enough
- Insist that vendors transfer knowledge, not just deliver systems
What This Means If You Are Hiring
The Saudi market now has a far larger pool of people who are AI-literate and a still-small pool of people who are AI-capable. Job descriptions that demand five years of LLM production experience are describing a candidate who mostly does not exist locally, at any salary.
The more effective pattern is to hire for engineering judgement and teach the AI specifics, while partnering for the parts you need working now. Capability that arrives with a delivery partner and stays with your team afterwards is worth more than either hiring or outsourcing alone.
Frequently Asked Questions
What is the SAMAI initiative?
SAMAI is Saudi Arabia’s national AI upskilling programme, which reported training more than one million participants in a single year as part of the Kingdom’s broader national strategy for data and AI.
Can short courses really create AI engineers?
No, and that is not their purpose. Mass programmes create fluency — a workforce that understands what AI can do, where it fails, and how to work with it. Deep specialists are built over years, not weeks.
Why do skills matter more than compute?
Compute and model licences can be purchased by anyone with a budget, so they are not durable advantages. The scarce input is people who can take AI systems into production and keep them working, which is why national training programmes matter strategically.
How should Saudi employers respond?
Raise general AI fluency across the organisation, protect and develop the small number of deep specialists you have, hire for engineering judgement rather than rare AI-specific experience, and require delivery partners to transfer capability to your team.
Conclusion
SAMAI is a recognition that the hardest part of national AI capability is human, not technical. Training more than a million people in a year does not produce a million engineers, but it does produce an economy that can absorb AI rather than merely purchase it.
For employers, the lesson is to stop competing only for scarce senior specialists and start building capability deliberately — internally where you can, with partners where you must, always insisting that the knowledge stays behind when the project ends.
How Elbetron Can Help
Services directly related to what you just read