KRI PRESS |
Jun 16, 2025
Malaysia’s role in AI Safety in Southeast Asia
2025 is shaping up to be the year in which AI governance takes a sharp turn towards spurring acceleration, departing from the global trend towards regulations and guardrails that was apparent in years prior.
A cluster of events early in the year instigated and marked this shift. As US President Donald Trump came into power in January, he signaled his support for Big Tech and revoked an executive order by the Biden administration that aimed to reduce and manage AI risks. Shortly after, Chinese company Deepseek created a “Sputnik moment” in AI as it unveiled its R1 model, assessed to be on par with US models, trained at a fraction of the cost and released for free.
In February, the AI Action Summit held in Paris broke from preceding summits that focused on AI safety concerns, to instead emphasise inclusivity and sustainability in AI development and adoption. During the event, the EU also committed to mobilise €200 billion of investment in AI, and dropped its proposal on the AI Liability Directive from the 2025 Work Programme, further indicating its move away from stricter regulations on the sector.
As countries and regions prioritise speed and jostle for technological supremacy, AI safety is sidelined. On top of that, as the world moves towards AI acceleration, geopolitical developments and trade wars threaten international cooperation that is foundational for global governance of AI.
What do these mean for us in Southeast Asia?
Southeast Asian countries need to work together to ensure AI safety
With eleven countries across varying stages of development, the region is home to Singapore which consistently leads in charts for government AI readiness and also countries such as Myanmar, Laos and Cambodia that are still working on digitalisation and internet connectivity.
The difference in AI readiness means that AI risks will be felt differently across different ASEAN member states (and also Timor Leste, which will soon become a full member). As elaborated within KRI’s recent report on AI governance, the level of technological readiness underlines a country’s ability to respond to three key concerns: (1) being left behind on AI adoption, (2) unsafe AI and unintended consequences, and (3) malicious use of AI.
However, beyond readiness in individual countries, it is in the interest of Southeast Asian countries to work together on uplifting the region’s readiness as a whole.
The primary reason for this is the transboundary nature of AI harms. For example, scam farms in Myanmar and Cambodia have quickly adopted AI to automate and scale online scams, causing billions of losses globally ($18-37 billion in 2023 alone). The harms are not only financial but also humanitarian. It has been estimated that more than two hundred thousand people from the region and beyond were trafficked and coerced into executing these scams under inhumane conditions.
Cooperation on capacity-building, incident monitoring and data sharing across all states would lead to safer AI for all. This would in turn increase public trust and literacy for more effective adoption as well as harmonised standards that would reduce compliance costs for businesses.
A united bloc with a combined population of 700 million will also stand a better chance in negotiations with larger players, such as stronger states and Big Tech companies. This ability to coordinate a coherent regional response has become a matter of significance in the era of Trump-induced trade wars and geopolitical instabilities.
Malaysia has a key role to play
With Malaysia assuming chairmanship of ASEAN in 2025, it has much to offer the region given the domestic AI policy journey it has undertaken.
This journey has been guided by the National AI Roadmap 2021-2025, which included establishing AI governance in Malaysia as a key strategy. Building upon the seven Responsible AI principles in AI-RMAP, Malaysia developed its first National Guidelines on AI Governance and Ethics. This document provides guidance and a code of ethics for AI developers, deployers and users to promote responsible AI in Malaysia.
With the establishment of the National AI Office (NAIO), Malaysia is one of a few ASEAN countries with a dedicated AI body. NAIO is the central authority in Malaysia tasked with advancing the nation’s AI policy and innovation agendas via seven main deliverables (see image below). To accelerate its progress towards these deliverables, NAIO formed multi-stakeholder Working Groups focusing on topics such as AI safety and engaged with industry players in sectors including healthcare and agriculture to explore the application of AI.
At the regional level, Malaysia is shaping the regional narrative by championing the establishment of an ASEAN AI Safety Network. This project, a key deliverable for Malaysia’s ASEAN Chairmanship, aims to create a central platform for ASEAN on AI safety. The Network would facilitate AI safety research, promote safe and responsible development and adoption of AI, and encourage harmonisation and interoperability of AI safety within ASEAN Member States. The Network would also enable ASEAN to have a unified voice and participate in global AI safety and governance discourses which are more often than not the domain of Global North countries.
Malaysia's AI journey has been supercharged in recent years with investments from both public and private sectors in AI infrastructure, AI talent and research as well as an emphasis on developing AI governance to enable the AI ecosystem. The country therefore is well positioned to play a key role in uplifting AI readiness of the region, not only in technical capacity but also in forming the thinking about AI’s impact on society in Southeast Asia.
Where Next?
Post-ASEAN Chairmanship, Malaysia needs to continue its momentum. Once NAIO’s foundation building efforts are completed domestically, it should continue championing AI safety at the regional level, leveraging the ASEAN AI Safety Network as a key platform.
Within the ASEAN sectoral bodies, Malaysia should also play a more active role in the ASEAN Digital Sector body. For example, via the ASEAN Working Group on AI Governance, Malaysia could champion programmes that support capacity building for ASEAN countries to operationalise the ASEAN Guide on AI Governance and Ethics as well as the Expanded ASEAN Guide on Gen AI.
AI’s rapid growth pushes countries to follow its tide or risk falling behind. As an emerging AI player, Malaysia is well-positioned to bring countries in Southeast Asia together to develop AI safety and governance paradigms suitable for the region. Recognising the region's varying AI readiness levels and socio-cultural nuances, Malaysia can lead efforts in this regard, allowing countries in the region to grow together, at their own pace.
Note: The authors contributed equally to this article
A cluster of events early in the year instigated and marked this shift. As US President Donald Trump came into power in January, he signaled his support for Big Tech and revoked an executive order by the Biden administration that aimed to reduce and manage AI risks. Shortly after, Chinese company Deepseek created a “Sputnik moment” in AI as it unveiled its R1 model, assessed to be on par with US models, trained at a fraction of the cost and released for free.
In February, the AI Action Summit held in Paris broke from preceding summits that focused on AI safety concerns, to instead emphasise inclusivity and sustainability in AI development and adoption. During the event, the EU also committed to mobilise €200 billion of investment in AI, and dropped its proposal on the AI Liability Directive from the 2025 Work Programme, further indicating its move away from stricter regulations on the sector.
As countries and regions prioritise speed and jostle for technological supremacy, AI safety is sidelined. On top of that, as the world moves towards AI acceleration, geopolitical developments and trade wars threaten international cooperation that is foundational for global governance of AI.
What do these mean for us in Southeast Asia?
Southeast Asian countries need to work together to ensure AI safety
With eleven countries across varying stages of development, the region is home to Singapore which consistently leads in charts for government AI readiness and also countries such as Myanmar, Laos and Cambodia that are still working on digitalisation and internet connectivity.
The difference in AI readiness means that AI risks will be felt differently across different ASEAN member states (and also Timor Leste, which will soon become a full member). As elaborated within KRI’s recent report on AI governance, the level of technological readiness underlines a country’s ability to respond to three key concerns: (1) being left behind on AI adoption, (2) unsafe AI and unintended consequences, and (3) malicious use of AI.
However, beyond readiness in individual countries, it is in the interest of Southeast Asian countries to work together on uplifting the region’s readiness as a whole.
The primary reason for this is the transboundary nature of AI harms. For example, scam farms in Myanmar and Cambodia have quickly adopted AI to automate and scale online scams, causing billions of losses globally ($18-37 billion in 2023 alone). The harms are not only financial but also humanitarian. It has been estimated that more than two hundred thousand people from the region and beyond were trafficked and coerced into executing these scams under inhumane conditions.
Cooperation on capacity-building, incident monitoring and data sharing across all states would lead to safer AI for all. This would in turn increase public trust and literacy for more effective adoption as well as harmonised standards that would reduce compliance costs for businesses.
A united bloc with a combined population of 700 million will also stand a better chance in negotiations with larger players, such as stronger states and Big Tech companies. This ability to coordinate a coherent regional response has become a matter of significance in the era of Trump-induced trade wars and geopolitical instabilities.
Malaysia has a key role to play
With Malaysia assuming chairmanship of ASEAN in 2025, it has much to offer the region given the domestic AI policy journey it has undertaken.
This journey has been guided by the National AI Roadmap 2021-2025, which included establishing AI governance in Malaysia as a key strategy. Building upon the seven Responsible AI principles in AI-RMAP, Malaysia developed its first National Guidelines on AI Governance and Ethics. This document provides guidance and a code of ethics for AI developers, deployers and users to promote responsible AI in Malaysia.
With the establishment of the National AI Office (NAIO), Malaysia is one of a few ASEAN countries with a dedicated AI body. NAIO is the central authority in Malaysia tasked with advancing the nation’s AI policy and innovation agendas via seven main deliverables (see image below). To accelerate its progress towards these deliverables, NAIO formed multi-stakeholder Working Groups focusing on topics such as AI safety and engaged with industry players in sectors including healthcare and agriculture to explore the application of AI.
At the regional level, Malaysia is shaping the regional narrative by championing the establishment of an ASEAN AI Safety Network. This project, a key deliverable for Malaysia’s ASEAN Chairmanship, aims to create a central platform for ASEAN on AI safety. The Network would facilitate AI safety research, promote safe and responsible development and adoption of AI, and encourage harmonisation and interoperability of AI safety within ASEAN Member States. The Network would also enable ASEAN to have a unified voice and participate in global AI safety and governance discourses which are more often than not the domain of Global North countries.
Malaysia's AI journey has been supercharged in recent years with investments from both public and private sectors in AI infrastructure, AI talent and research as well as an emphasis on developing AI governance to enable the AI ecosystem. The country therefore is well positioned to play a key role in uplifting AI readiness of the region, not only in technical capacity but also in forming the thinking about AI’s impact on society in Southeast Asia.
Where Next?
Post-ASEAN Chairmanship, Malaysia needs to continue its momentum. Once NAIO’s foundation building efforts are completed domestically, it should continue championing AI safety at the regional level, leveraging the ASEAN AI Safety Network as a key platform.
Within the ASEAN sectoral bodies, Malaysia should also play a more active role in the ASEAN Digital Sector body. For example, via the ASEAN Working Group on AI Governance, Malaysia could champion programmes that support capacity building for ASEAN countries to operationalise the ASEAN Guide on AI Governance and Ethics as well as the Expanded ASEAN Guide on Gen AI.
AI’s rapid growth pushes countries to follow its tide or risk falling behind. As an emerging AI player, Malaysia is well-positioned to bring countries in Southeast Asia together to develop AI safety and governance paradigms suitable for the region. Recognising the region's varying AI readiness levels and socio-cultural nuances, Malaysia can lead efforts in this regard, allowing countries in the region to grow together, at their own pace.
Note: The authors contributed equally to this article
related to this event
Want more stories like these in your inbox?
