Malaysia wants to crack the world's top 10 AI nations by 2030, and its National AI Office boss says the real challenge is skilling up an entire nation

Interest.co.nz is unpacking how AI will change your everyday life - the risks, the opportunities, and what to actually expect. Our new series brings you the policymakers, experts and industry leaders from New Zealand and overseas.

By Anna Whyte - This article was made with help from funding from the Asia New Zealand Foundation.

Malaysia wants to be a global leader when it comes to AI.

In an attempt to speed up AI adoption and ethical development of artificial intelligence, Malaysia launched its National AI Office (NAIO) in 2024. Interest.co.nz travelled to Kuala Lumpur and sat down with NAIO chief executive Sam Majid to ask how he plans to do it. 

The AI building block

NAIO was Majid's third 'start-up', of sorts.

In 2024, he was a one-man band with a hefty KPI (Key Performance Indicator) of seven tasks to finish off within a year. Those tasks included developing an AI technology plan, which would become Malaysia's AI plan of what it should be doing when it comes to AI across 28 different sectors.

He also needed to roll out a regulatory framework, promote the widespread adoption of AI, a code of ethics, assess the impacts of AI across Government, and produce a market overview on AI trends and datasets on AI technology.

Majid said when AI became a mainstream conversation, it was important Malaysia had a focal point on how the country should respond to AI development, and stay close to the impact AI will have - both positive and negative.

Part of that was through understanding how other jurisdictions safeguard their citizens.

Investing in AI

Addressing digital changes and staying ahead of the AI curve - or at least trying to - is essential, Majid said.

"AI is powered by investment. You don't need to be number one, you just need to be very clear on what sort of future that your country should have - powered by AI."

"It’s a building block," Majid said.

"So in the space of AI, the building blocks mean that investors come from all walks of life, they invest in data centres, they invest in people, they invest in models, they invest in use cases. All of them want to see a few things that should be present in the country, like, for example, does it have a stable administration? Does it have a cohesive framework and blueprint that matters to the economy... so does the country know what to do when it comes to AI? So, if the answer is yes, yes, yes, yes, then investors feel that instead of putting some money into other countries that are maybe unsure.

"They are better off putting [their investments] in a country that is very sure of how to drive future growth."

How to roll out an AI strategy

Majid said the number one piece of advice when it comes to a country's AI strategy would be to make sure it's a whole-of-nation approach.

"This is key, because AI is one of those tools that impacts every walk of life, whether you are a student, CEO, public servant, enterprise, it affects you in some way."

He said Malaysia's AI plan "gives a lot of confidence to everybody that we are all together on the same journey".

And it has got a goal as well. That goal included having specific KPIs for the number of scientists, engineers, AI champion companies and AI patents.

The nation's playbook also aims to get Malaysia up into the top 10 of global AI rankings by 2030, "which means it's not just investment, it's also focused on talent", Majid said.

"The second part is don't forget your biggest sectors... the government sector, the healthcare sector, and the other key sectors..."

AI in the workspace

How do you build up an AI competent workforce? That question is front and centre for Majid.

Talent is not only the biggest, but the most challenging element of Malaysia's AI plan.

They have broken it down into three areas. "Number one, existing workforce, people like you and I... we need to know where we will be in five years' time now that AI is in the workspace, so we need to be set up for upskilling, retraining.

"The second group of talent is the incoming workforce. The ones that are finishing universities or in the higher education space. It's not good enough that our universities are still offering courses and subjects that are redundant in some way.

"The third group of talent is the one that are in kindergarten to Year 12. In 10 years' time they'll be the next workforce. They have to learn how to use AI responsibly; teachers need to be in classes teaching AI side by side."

He said a lot of people were still unaware of what AI can do.

"Most of those that are a little bit behind are board members of companies... the tone from the top, the tone from the top of government, tone for the top of enterprises, is key."

How do you actually roll it out?

Generative AI (Gen AI) tools were quickly rolled out across the public sectors, close to half a million. But access isn't good enough, change management and training are needed too, Majid said.

"We monitor the uptake, we monitor what sort of use case the public servant, civil servant have been using Gen AI for. For example, use case of taking minutes of meetings, use case of saving time in producing policies, use case of automation in terms of processes."

For AI implementation, he said it typically started with automating repetitive tasks.

"The second one, especially due to agentic AI, is where AI is being used to rewire processes", that can then become autonomous," he said.

"The third one... is AI innovation, where AI is being used without any legacy or baggage, that means it's a company producing a brand new product, no legacy of past processes, completely built from the ground up using AI."

Majid said innovation is where the focus needed to be as well.

In Malaysia, the conversation about how much money is saved in the public sector through AI is equated to a productivity conversation, he said.

"Malaysia needs to do many interventions when it comes to the AI investment in public sector. If those interventions are not performed, then it will be a paradox, meaning money [is] spent but productivity isn't there."

What does an AI Nation look like in 2030?

Majid said the rollout of Malaysia's AI plan required a demand-led rollout, communication and influencing.

Majid said if the impact of AI is high, then they would want an AI ecosystem to be governed by local laws, which means the data stays in Malaysia, as well as the infrastructure and the legal infrastructure.

"Malaysia shouldn't be just a consumer nation of AI... we should have adequate champions that can build [a] made-by Malaysia AI environment, and this is key, because not every AI is being used in a simple way. Some AI use cases are high risk."