Malta is giving citizens free ChatGPT Plus.
But only after training.

The headline sounds like free ChatGPT Plus for an entire country. But the real hook is the order.

Malta is not simply handing citizens premium AI access and hoping they figure it out. On 16 May 2026, Malta’s government launched AI for Everyone in partnership with OpenAI and Microsoft: a free national AI literacy programme open to every citizen and resident aged 14 and over. Complete the course. Earn a certificate. Then get a free year of ChatGPT Plus or Microsoft 365 Copilot.

Not before training. After training.

That is the part Malaysia should pay attention to. Malta made AI access the reward for learning, not the starting point. The country is effectively saying: before we give you powerful AI tools, we want you to understand how to use them responsibly, practically, and productively.

Most organisations do the opposite. They buy ChatGPT, Copilot, or other AI licences first, then hope employees will somehow build the skills later. Malta flipped that model. Training comes first. Access comes second. Capability becomes the gateway.

Malaysia is building one of the most ambitious national AI strategies in Southeast Asia. It has the investment, the policy, and the political will. The next step is applying the Malta principle at scale: structured AI training before mass AI access, especially inside the private sector workforce.

Why the sequence is everything

The Malta programme isn’t complicated in structure. Three core modules covering AI fundamentals, practical application, and ethics. Roughly two hours of content. Fully online, self-paced, accessible from any device using a government eID. The University of Malta built the educational content. MDIA runs the programme. OpenAI and Microsoft brought the infrastructure.

What makes it significant isn’t the length. It’s what the design refuses to do.

It refuses to give anyone the tool before they’ve earned it through training. The ChatGPT Plus subscription isn’t a perk for signing up. It’s contingent on finishing. That single design decision changes the entire nature of what the programme produces: not a population with AI access, but a population with a structured reason to understand AI before they use it.

Most national digital initiatives get this backwards. They fund access and hope literacy follows. Malta inverted the model entirely. And that inversion is precisely what Malaysian organisations — not just the government, but every HR leader, every L&D manager, every CEO buying AI licences — need to understand.

The numbers shaping Malaysia’s moment

The opportunity is clear, but so is the gap. Malaysia is already investing heavily in AI readiness, yet the workforce is moving faster than structured training can reach it.

At the national level, the ambition is real. Malaysia has committed RM16.36 billion to the Ministry of Digital for strategic AI initiatives, launched programmes such as the National AI Talent Accelerator and Jelajah AI MyMahir, and positioned the 2026 to 2030 National AI Action Plan around building Malaysia into a regional AI hub.

But policy ambition and workforce capability are not the same thing. Government programmes can create awareness and reach selected groups, while the private sector workforce is already using AI at scale. The gap between 93% usage and 12% sufficient training is where the urgency sits: employees have access, but most do not yet have the structure to turn it into real capability.

The same mistake, repeated at scale

Here’s what most corporate AI rollouts look like. An organisation buys a block of Copilot or ChatGPT licences. Teams get access. Someone sends a getting-started guide in the company chat. Maybe there’s a lunch-and-learn. That’s the training programme.

The result is exactly what Malaysia’s numbers show. Usage is high, but capability remains shallow. That gap isn’t a coincidence. It’s the predictable outcome of giving access before building understanding.

AI awareness and AI capability are not the same thing.

Confusing the two is how organisations end up with widespread usage but shallow capability.

Malta’s government started from a different assumption: that access without training is a wasted resource. That handing someone a premium AI subscription before they understand what AI is produces very little beyond a subscription fee. So they built the prerequisite first. The tool comes after you’ve earned it.

That’s the model. And it’s the model Malaysian organisations need to apply internally, whether a government mandate exists or not.

The gap that awareness doesn’t close

There’s a version of the AI training conversation that relies entirely on individual initiative. Motivated employees will find the resources. Curious professionals will watch the tutorials. The ambitious ones will complete a free online course on their own time.

That version isn’t wrong. But it produces a small group of confident users while the rest of the workforce is left to improvise.

Unstructured, self-directed AI exposure doesn’t reliably build applied capability. It builds familiarity with the easy tasks and avoidance of the harder ones. It creates a quiet split inside organisations that haven’t noticed it yet: the employees who figured it out, and the ones quietly falling behind.

56% of working professionals in Malaysia use AI tools at work. Only 26% have received any formal training to do so. The gap between using a tool and using it well is exactly where productivity gains disappear. It’s also where the 697,000 workers most exposed to AI disruption are sitting right now, without a structural path forward.

The workforce isn’t resisting AI. It’s using AI without a map.

That’s a different problem, and it needs a different solution.

Training first. Everything else after.

Malaysia has done the awareness phase well. RM16.36 billion into the Ministry of Digital. A national AI action plan. Government programmes reaching students, civil servants, and job seekers. The narrative is clear and the political will is real.

What awareness doesn’t produce, on its own, is a workforce that can execute. That requires structured training designed around how people actually work — not a licence, not a video, not a certificate for watching a module. Applied training. Assessed capability. A clear sequence where understanding comes before access, and progress is measured by output rather than completion rate.

That’s what Malta built for a nation. It’s what Malaysian organisations need to build for their teams.

60% of Malaysian employers believe their organisation needs AI to stay competitive. Over half aren’t confident they can execute it. That gap doesn’t close with another government programme. It closes when employers make the same decision Malta’s government made: training is the prerequisite, not the afterthought.

This is the gap SkillTrainer AI was built to close. An AI training platform built in Malaysia for organisations that want to apply the Malta principle internally: train first, unlock capability after. Teams learn AI through structured pathways, applied tasks, and continuous assessment before AI becomes another unmanaged tool in the workplace. The goal isn’t completion. It’s a workforce that reaches for AI automatically, because they know what to do with it.

Malta proved the sequence works at a national level. The organisations that apply it at the team level in 2026 will be running a measurably different workforce by 2027.