Malaysia unveils bold AI bill, eyes 300,000 new jobs by 2030

Malaysia’s AI governance strategy is anchored in a comprehensive two-pronged legal approach that enforces existing laws against content misuse while introducing a pioneering AI Governance Bill. This bill specifically targets early risk mitigation for advanced AI threats such as deepfakes and synthetic content, balancing innovation with public safety and individual rights. The government’s commitment extends to combating child exploitation and unlawful content through stringent data protection measures and product assessments, thereby fostering a secure and accountable AI ecosystem.

The nation is advancing a risk-based, evidence-driven AI governance framework that emphasizes system security, data protection, and digital sovereignty regardless of an AI platform’s origin, as exemplified by oversight of platforms like DeepSeek. Institutional mechanisms such as the Malaysian AI Safety Institute, a National AI Action Plan, and a dedicated national AI security committee are being established to ensure transparent and accountable AI use. Public sector AI deployment is subject to strict compliance policies to safeguard sensitive information, reflecting Malaysia’s holistic approach to governance.

Malaysia’s AI Governance Bill, currently undergoing public consultation, adopts a principle- and risk-based framework that classifies AI systems by their risk level to tailor regulatory oversight. It mandates AI incident reporting and establishes a central AI authority to coordinate enforcement alongside sector-specific regulators, focusing on transparency, accountability, and responsible data governance. Notably, the bill excludes personal use and national security matters, signaling a targeted regulatory scope aimed at commercial and public AI applications.

Recognizing that security must be embedded from the earliest design stages, Malaysian leaders including IBM’s Sourabh Mallick advocate for integrating security into AI planning rather than retrofitting it post-development. This approach is complemented by the Data Sharing Act 2025, which mandates inter-agency agreements on data sharing to overcome historical silos, and modernization efforts that wrap legacy systems with APIs to enable secure, need-to-know data exchange. Malaysia’s digital sovereignty prioritizes control over data security rather than physical location, supporting hybrid storage models fortified with quantum-safe cryptography and crypto-agility strategies. This comprehensive governance vision is framed as a strategic, cross-agency cultural shift beyond mere IT initiatives, ensuring Malaysia’s AI ecosystem is both innovative and trustworthy.

Malaysia’s commitment to securing sovereign AI infrastructure is underscored by vigilant government scrutiny of AI platforms like DeepSeek, which faced criticism for storing personal data in China and thus risking exposure to foreign intelligence laws. To safeguard sensitive information, especially within the public sector, Malaysia mandates strict compliance with policies that localize data workloads and maintain control over AI systems, reinforcing national data sovereignty and security.

Recognizing that security is often an afterthought, Malaysian government and IBM leaders emphasize embedding security from the earliest design stages of AI systems to build trusted and interoperable government AI ecosystems. This approach is complemented by a pragmatic hybrid data sovereignty model, as Mohd Firdaus Bin Mohamed Khairi explains, where not all data must reside domestically if robust security measures are in place, balancing control with operational flexibility.

Malaysia’s AI Nation 2030 strategy aims to transcend basic data center infrastructure by fostering sovereign and localized AI ecosystems that enhance compute access, governed national datasets, and proprietary AI models. This plan also positions the government as an anchor customer to generate initial datasets and use cases, thereby catalyzing commercialization of domestic AI solutions while ensuring data sovereignty through foundational enablers like high-performance computing and AI governance frameworks.

Strategic partnerships with regional players such as Japan’s SoftBank Telecom, Singapore’s SC ZEUS, and Malaysia’s Robust HPC Group highlight Malaysia’s investment in cutting-edge AI data center infrastructure featuring liquid cooling and 800 VDC architecture to support scalable, real-time AI workloads. These efforts, coupled with Selangor’s pioneering RM2 billion sovereign digital infrastructure prototype and policies mandating at least 30% local content in data centers, underscore a robust commitment to localizing AI workloads and protecting national interests from cyber threats and foreign cloud dependencies.

Despite advanced and secure storage and processing layers, Malaysia’s sovereign AI cloud infrastructure faces a critical vulnerability at the data ingestion layer, where human-verified inputs—often influenced by corporate KPIs rather than objective truth—risk poisoning AI systems with sanitized or inaccurate data. To address this, experts advocate deploying deterministic hardware gateways at the physical edge to capture data directly from source machines, effectively removing the human middleman and ensuring data integrity essential for true sovereignty and trustworthy AI outcomes.

Malaysia’s strategic approach to workforce development in AI centers on targeted reskilling and upskilling initiatives aligned with sector-specific needs, as emphasized by Digital Minister Gobind Singh Deo who highlighted the identification of vulnerable sectors and the collaboration with universities and training providers to embed practical AI knowledge. This is further reinforced by the National AI Office’s role in coordinating federal and local governance frameworks to ensure AI development aligns with national security and sectoral regulations, creating a cohesive ecosystem for ethical and scalable AI integration.

Despite high experimentation rates—such as 71% of Malaysian banks adopting AI—full-scale strategic deployment remains limited at 17%, hindered by talent shortages, legacy systems, and governance gaps, underscoring structural challenges to responsible AI scaling. Government entities face parallel hurdles, grappling with data security concerns and fragmented, bottom-up AI pilots that yield low ROI; as Beth Howen notes, 95% of government AI use cases fail to deliver returns due to lack of coordinated strategy, highlighting the urgent need for guided roadmaps that prioritize impactful, ethical AI initiatives aligned with public accountability.

The AI Nation 2030 initiative crystallizes Malaysia’s commitment to workforce and ecosystem development by targeting the creation of 300,000 AI-related jobs through foundational enablers like a Global Talent Network and AI Growth Zones, which serve as catalytic platforms for accelerating sector-specific AI adoption. This comprehensive framework addresses diverse industries via 14 Impact Engines, ranging from healthcare to MSMEs, while positioning the government as an anchor customer through integrated AI applications in public services—such as MyGOV Malaysia and AI-assisted civil service workflows—to stimulate domestic AI innovation and workforce engagement.

At the local level, Selangor exemplifies coordinated AI adoption and workforce development by mandating councils to optimize 25 internal workflows with AI, aiming for at least 50% reductions in turnaround times and 30% cost savings. This initiative is bolstered by the appointment of specialized AI analytics consultants to dismantle entrenched data silos and the integration with MCMC’s RM2 billion sovereign digital infrastructure prototype, which provides a unified, secure technical stack for scalable AI deployment. Complementing these technological advances, Selangor’s strategic selection of AI training providers ensures civil servants are equipped to navigate and sustain AI integration effectively.

Building public trust in Malaysia's AI ecosystem hinges on transparent governance that addresses algorithmic biases and digital echo chambers, as highlighted by Mediha Mahmood, CEO of Content Forum Malaysia, who warns against AI-driven platforms that perpetuate repetitive content through autoplay and endless scrolling. This challenge transcends national borders, necessitating regional cooperation within ASEAN, where representatives like Hazremi Hamid emphasize the importance of shared AI governance principles that accommodate varying levels of digital maturity across member states.

Malaysia is pioneering a risk-based governance framework to balance rapid AI innovation with ethical safeguards, exemplified by initiatives such as the AI Governance Bill, the National AI Office, and the AI Safety Institute. This approach is complemented by a call from Aida Azmi, joint managing director of Pemandu Associates, to treat AI policies as evolving products that adapt iteratively through real-world feedback, ensuring regulations remain flexible and responsive amid fast-paced technological advancements.

AI is rapidly evolving in Malaysia from a mere productivity tool into a strategic asset for climate-risk management, particularly as climate risks increasingly impact corporate earnings and asset valuations. BIMB highlights that institutional investors are adopting AI-driven climate analytics to integrate these risks into portfolio management, mirroring global best practices. This transformation is underpinned by the convergence of national frameworks like the Thirteenth Malaysia Plan and the National Energy Transition Roadmap (NETR), which is projected to mobilize over MYR 600 billion by 2050, fueling demand for AI solutions across renewable energy, smart grids, and climate adaptation sectors.

The AI Nation 2030 initiative marks a strategic pivot for Malaysia, aiming to transcend its role as a data center hub by cultivating a value-driven AI ecosystem. This plan targets 14 sector-specific Impact Engines—including healthcare, smart cities, and public services—supported by foundational enablers like talent development, AI governance, and high-performance computing. Early catalytic projects focus on healthcare, AI cities, and public services, with government platforms such as MyGOV Malaysia poised to act as anchor customers, thereby accelerating AI adoption and generating critical datasets for scalable applications.

Selangor is pioneering Malaysia’s ambition to become an AI-driven smart state by deploying a RM2 billion sovereign digital infrastructure prototype developed by MCMC, designed to ensure data sovereignty and cybersecurity for critical public services. The state mandates that local councils optimize 25 internal workflows via AI agents, aiming for a 50% reduction in turnaround times and 30% cost savings, while also enforcing a policy requiring at least 30% local content in data centers to stimulate domestic semiconductor and engineering sectors. Complementing these efforts, Selangor is advancing high-granularity digital twins—3D virtual city models capturing topography and drainage—to enhance climate resilience and flood mitigation.

AI Nation 2030 also emphasizes sustainable infrastructure growth by linking data center incentives to renewable energy and power efficiency targets, supporting AI-enabled management of renewable resources and flexible grids. This approach aligns with the broader NETR goals and reflects Malaysia’s commitment to integrating AI solutions that not only drive economic growth but also bolster environmental sustainability through advanced analytics and smart energy systems.