CYBERJAYA (July 28): AI Malaysia Bhd has been launched as the country’s national artificial intelligence (AI) entity under the Ministry of Digital, formally institutionalising the National AI Office (NAIO).

NAIO, which was established on Dec 12, 2024, initially served as the government’s coordinating office for AI. Its transition into AI Malaysia gives the entity a permanent structure, broader responsibilities and the mandate to implement the country’s AI agenda.

AI Malaysia was established to provide a permanent institution to lead and coordinate the implementation of the government’s AI policies across ministries, agencies and other stakeholders.

The leadership of the new entity has yet to be announced. Shamsul Izhan Abdul Majid, who had headed NAIO since its establishment, ended his tenure in June.

The government also unveiled the National AI Action Plan 2026-2030, which will serve as the implementation framework for developing Malaysia’s AI ecosystem and supporting the country’s ambition to become an AI Nation by 2030.

The plan sets out 14 initiatives focused on applying AI in specific sectors and another 14 aimed at building the talent, innovation, infrastructure, governance and financing needed to support them.

AI Malaysia will be the central body linking national AI policies with their implementation across ministries and government agencies. It will also monitor the progress and impact of initiatives under the action plan.

Prime Minister Datuk Seri Anwar Ibrahim, who officiated the launch on Tuesday, said AI would require changes across education, government and the private sector. However, access to the technology must not be limited to urban communities or economic elites, he said.

“The challenge for the Digital Ministry is to ensure that this change does not widen the gap between urban and rural communities, or between the elite and the poor, and to ensure from the outset that it is more equitable,” Anwar said.

“We are proud that, despite our problems and constraints, Malaysia has begun moving forward at a faster and more convincing pace… No country can build what is called an AI Nation at the old pace. The changes are unprecedented and require a speed that we have never achieved before.”

Anwar also warned of risks relating to national security, sovereign cloud arrangements, the reach of the US Cloud Act and cybersecurity. He raised concerns about the power of large global technology companies.

He said Malaysia could not reject AI but needed to understand the technology, manage its risks and build its own capabilities.

“In the debate over AI, we cannot begin from an anti-technology position. We must first master it. The danger is real, there are large companies that want to dominate the world through their AI capabilities and expertise. Does that mean we should abandon AI altogether? That is impossible. We must master the knowledge,” he said.

“The entire education system, beginning at the primary level, must provide exposure to AI. Teachers must be trained, additional funding provided and basic infrastructure overhauled.”

Anwar announced that the AI Untuk Rakyat programme would open on Aug 31. Under the programme, people aged between 18 and 30 who complete specified course modules under Rakyat Digital will receive a free three-month subscription to a leading AI application.

The programme is expected to benefit 100,000 young people.

Digital Minister Gobind Singh Deo, who was also at the launch, said AI Malaysia would provide the institutional structure needed to lead and coordinate the implementation of Malaysia’s AI Nation ambition.

“We will ensure that we take action to build an ecosystem that is ready to be used so that people can use this technology. And the most important thing is for us to ensure that we can build a safe ecosystem so that this new technology that we will use can be used for our benefit and at the same time our safety is ensured,” he said.

“We have less than 60 months to fulfil the vision of AI Malaysia 2030.”

AI Malaysia’s work will focus on developing local AI talent, strengthening the country’s data and computing infrastructure, and accelerating AI adoption across the public sector and industries.

The institution will also coordinate Malaysia’s international AI partnerships and develop governance frameworks, ethical standards and other mechanisms intended to strengthen public trust in the technology.

In addition, the Malaysian AI Safety Institute will be established under AI Malaysia to conduct safety assessments, test AI models and undertake AI red-teaming, where systems are deliberately challenged to uncover weaknesses and potential harms.

The government is also drafting an AI governance bill that adopts a risk-based approach. The proposed legislation is open for public consultation and is targeted for completion by the end of 2026.