AI in Malaysia: Governance, Adoption and the Path Forward

Artificial intelligence (AI) is revolutionising industries across Malaysia, unlocking unprecedented opportunities for efficiency, automation and economic growth. By 2030, AI is expected to contribute USD115 billion to Malaysia’s productive capacity. Recognising AI’s transformative potential, Malaysia’s Prime Minister Anwar Ibrahim emphasised: “If you want to ensure that an emerging economy succeeds, remains competitive, and sustainable, then it has to be through a quantum leap, and AI is the answer for that.” This commitment is reflected in Malaysia’s 2025 budget, which allocates MYR600 million for AI research and development and MYR50 million to expand AI-related education, underscoring the government’s strategic push toward an AI-powered future.

AI Governance in Malaysia

There is currently no specific legislation regulating the use of machine learning and AI in Malaysia.

National guidelines on AI governance and ethics

The Ministry of Science, Technology and Innovation (MOSTI) has introduced the National Guidelines on AI Governance and Ethics (the “AI Guidelines”), a non-legally binding framework to provide guidance on this subject.

The AI Guidelines primarily aim to promote responsible and ethical AI development and deployment across sectors and support the implementation of the Malaysia National AI Roadmap 2021–2025. The AI Guidelines are designed for three main stakeholders:

• AI end users;
• policy makers in the government, agencies, organisations and institutions; and
• developers, designers, technology providers and suppliers.

Serving as a voluntary and non-binding set of recommended practices, the AI Guidelines outline seven core principles (the “Guidelines’ Seven Principles”), as follows.

• Fairness – AI should be free from bias and ensure equitable outcomes.
• Reliability, safety and control – AI must function as intended and allow human intervention when necessary.
• Privacy and security – AI systems must safeguard personal data and prevent misuse.
• Inclusiveness – AI should be designed to benefit diverse populations without discrimination.
• Transparency – AI decision-making processes should be explainable and understandable.
• Accountability – AI developers and users must take responsibility for the outcomes of AI systems.
• Pursuit of human benefit and happiness – AI should enhance human well-being and contribute positively to society.

These principles align Malaysia with global best practices, such as the OECD AI Principles and Singapore’s Model AI Governance Framework. While these AI Guidelines represent a significant step forward, Malaysia’s legal AI framework remains in its early stages.

Launch of National AI Office

To address the nascent nature of AI governance in Malaysia, the government launched the National AI Office (NAIO) on 12 December 2024, responsible for shaping AI policies, governance and investment strategies. NAIO’s initiatives include:

• AI Technology Action Plan 2026–2030;
• AI Adoption Regulatory Framework;
• AI Code of Ethics; and
• Public Sector AI Adaption Guidelines, which have begun scaling AI use within government agencies, including the rollout of Google Workspace’s Gemini Suite to 445,000 public officers.

While NAIO’s formation signals progress, Malaysia’s AI regulatory framework remains fragmented. For example, the AI Guidelines emphasise transparency, but Malaysia’s Personal Data Protection Act 2010 (“MY PDPA”) does not yet regulate automated decision-making (ADM) – an area that is increasingly relevant as AI adoption grows. In contrast, the EU General Data Protection Regulation (EU GDPR) grants individuals rights against ADM-based decisions, setting a benchmark for future enhancements in Malaysia’s regulatory approach.

AI Adoption in Malaysia: Opportunities and Challenges

Business and economic impact

AI is reshaping Malaysia’s business landscape, boosting productivity in small medium enterprises (SMEs) and enhancing efficiency in large corporations. According to the Malaysian Digital Economy Corporation (MDEC), 140 AI solution providers have collectively generated MYR1 billion in revenue, demonstrating Malaysia’s potential as a regional digital economy leader.

However, the country faces a significant AI talent shortage. A 2024 Amazon Web Services report found that 81% of Malaysian employers struggled to hire AI talent, despite 90% prioritising AI skills. The World Bank estimates that Malaysia has only 3,000 AI professionals, while demand is expected to reach 30,000 by 2030.

To address this, public-private collaborations are ramping up AI education. For instance, Microsoft’s AI for Malaysia’s Future (“AIForMYFuture”) initiative aims to train 800,000 Malaysians by 2025 through online modules and hands-on workshops. Through AIForMYFuture, Microsoft will collaborate with stakeholders including government, industry, education and civil society to develop AI skills at all levels of society.

Beyond boosting productivity and economic growth, AI is also transforming key industries. From finance and healthcare to agriculture and smart cities, AI is driving innovation and improving efficiency across various sectors.

Agriculture

Malaysia is leveraging AI to modernise agriculture, with Malaysia launching the world’s first AI-driven palm oil mill. The mill uses predictive analytics, automation and real-time data monitoring to boost productivity while reducing environmental impact and reliance on foreign labour.

Malaysia is the second-largest palm oil producer globally, and adopting this technology across the nation’s oil mills could reduce foreign labour dependency by up to 35%.

Additionally, Rakan Tani, a groundbreaking digital platform, uses AI-powered order matching to help farmers secure buyers early in the crop cycle. This helps farmers receive competitive pricing based on their projected yields, promoting predictability and financial stability.

Medical and healthcare

AI is transforming Malaysia’s healthcare sector, particularly in early disease detection. The Lung Cancer Network of Malaysia (LCNM), in collaboration with AstraZeneca Malaysia, successfully diagnosed and treated Malaysia’s first lung cancer case detected using AI. This milestone builds on the growing use of AI-assisted screenings, which have demonstrated higher accuracy than traditional methods in detecting lung abnormalities, leading to earlier and more effective treatments.

Similarly, the Ministry of Health unveiled “DR. MATA”, an AI-powered diagnostic tool to detect diabetic retinopathy, a leading cause of blindness among diabetics. This demonstrates the healthcare industry’s keenness to transition to a proactive and preventative model, aiming to cultivate a healthier, more informed population.

AI is also digitally transforming the operations of public health clinics, with the roll-out of cloud-based digital management system (CCMS), enhancing efficiency and accessibility of healthcare services. AI tools like chatbots and virtual assistants are also being deployed to help ease the workload of healthcare workers – a long-standing challenge in Malaysia’s public healthcare system.

Smart cities and transportation

AI is a key driver in Malaysia’s Smart City Framework, a national level framework serving as a guide to local authorities in planning and developing smart cities in Malaysia. The government views smart cities as the future of urban planning, development and management, leveraging technology to address challenges such as inefficient public services, environmental pollution and traffic congestion.

One key area of AI adoption is traffic management. In Kuala Lumpur, a citywide CCTV network enables AI platforms to count vehicles, classify them by type, and recognise licence plates. This allows real-time traffic analysis, offering insights into road load, peak hours and traffic patterns. From the data, authorities are able to optimise traffic flow, plan infrastructure and make data-driven decisions for smarter urban mobility.

AI is also transforming transport management, with the Road Transport Department (JPJ) planning to use AI-driven systems to detect and prevent traffic offences. These systems aim to streamline the processing of violation reports, enabling faster response times and improving road safety.

Financial services

In the financial services sector, AI adoption is widespread amongst financial service providers (FSPs), particularly in fraud detection. FSPs are deploying AI and machine learning in credit underwriting, anti-money laundering and fraud detection, electronic know-your-customer (e-KYC), customer analytics, trading, customer engagement and technology risk management.

In August 2024, the Central Bank of Malaysia, in partnership with PayNet, and other financial institutions, launched the National Fraud Portal (NFP), utilising AI to combat financial fraud through predictive analysis. The system enables financial institutions and the National Scam Response Centre (NSRC) to swiftly identify, trace and freeze suspicious transactions. As a result, the time taken to trace stolen funds has been reduced by 75%, from two hours to merely 30 minutes.

AI and Personal Data

The increasing integration of AI in decision-making processes has raised important concerns regarding data privacy, transparency and accountability. AI systems, particularly those using ADM, rely on significant amounts of personal data to generate insights, assess risks and make real-time decisions. However, the complexity of AI models, often referred to as the “black box” problem, makes it difficult to interpret how AI-driven decisions are made – posing significant challenges for data protection and consumer rights. 

Regulation of automated decision-making

A key issue in Malaysia’s regulatory landscape is that the MY PDPA does not currently regulate ADM. This contrasts with the EU GDPR which grants individuals the express right to not be subject to decisions based solely on automated processing, including profiling, unless certain conditions apply.

Without similar safeguards in Malaysia’s legal framework, individuals may lack clarity and recourse when AI systems are used to determine outcomes in areas such as credit scoring, insurance underwriting and employment screening. For example, in the banking sector, AI is widely used for credit risk assessments, where an applicant’s eligibility for a loan may be determined by an AI algorithm without human intervention. Without clear regulatory requirements on explainability and oversight, consumers may find it difficult to challenge decisions that affect their financial standing.

Principle of transparency in AI

Recognising these gaps, Malaysia has taken initial steps towards strengthening AI governance. The AI Guidelines advocate for transparency in AI systems with one of the Guidelines’ Seven Principles being Transparency. This principle of Transparency posits that any capabilities of an AI system used in decision-making processes should be explainable, including both the technical processes and the related human decisions.

This level of transparency enables stakeholders to evaluate any perceived risks related to AI and address any related concerns. The AI Guidelines outline five key elements that should be adhered to in promoting transparent decision-making:

• full disclosure about the fact that an AI system is being used in decision-making;
• the intended purpose of the AI systems;
• the training data (describing the data that was used in training, what historical and social biases exist in the data and what procedures were used to verify data quality);
• maintenance and assessment of AI systems; and
• the ability to challenge the decisions made by the AI systems.

Future introduction of profiling and decision-making guidelines

In addition to the voluntary AI Guidelines, the Malaysian government is working towards more concrete regulations. The Profiling and Decision-Making Guidelines, expected to be introduced soon, are part of a series of seven guidelines to be introduced pursuant to the MY PDPA. The Digital Minister has acknowledged the urgent need for clearer rules in this area, signalling that the Profiling and Decision-Making Guidelines will likely incorporate AI-specific provisions, including regulating ADM. In recent times, two of the seven anticipated guidelines have been issued, namely the Notification of Data Breach Guidelines and the Data Protection Officer Guidelines.

As Malaysia continues to expand its AI ecosystem, the development of a robust legal and ethical framework will be crucial to ensuring responsible AI deployment. Future reforms may need to consider stricter obligations for AI-driven ADM, enhanced consumer rights protections and industry-specific regulations for high-impact sectors such as financial services, healthcare and government services.

AI and intellectual property

The rapid advancement of AI-generated content has raised complex questions about the intersection of artificial intelligence and intellectual property (IP) law. As AI systems become increasingly capable of producing text, images, music and even inventions, traditional IP frameworks – which are built on the notion of human authorship and ownership – are being challenged. For example, in traditional copyright, the creator of a work typically assumes ownership. However, AI blurs the lines of ownership where the question arises if authorship should rest with the human programmer who designed the AI system, the individual providing input or prompts to the AI, or whether AI entities can claim copyright for autonomously generated content.

As AI technology advances it raises pressing questions about the applicability of traditional IP laws to works generated by AI. A highly debated topic that has taken centre stage in global discourse is whether using copyright-protected materials to train AI models constitutes an infringement of the original copyright. Alternatively, there is the question of whether AI models should be granted the right to generate new content based on information derived from their training data. Several high-profile lawsuits have been filed worldwide against AI developers, alleging that their models have been trained on copyrighted material without consent, leading to potential violations of copyright and fair use principles.

AI and copyright law

In Malaysia, the Copyright Act 1987 does not explicitly address AI-generated works, raising critical questions about ownership, originality and infringement in the digital age. Under current law, the language used in Section 10 of the Copyright Act provides that copyright subsists in every work eligible for copyright of which the author, or in the case of joint ownership, any of the authors, is, at the time when the work is made, a “qualified person”.

The Copyright Act defines “qualified person” to mean:

• in relation to an individual, a person who is a citizen of, or a permanent resident in, Malaysia; and
• in relation to a body corporate, a body corporate established in Malaysia and constituted or vested with legal personality under the laws of Malaysia.

While this notion of ownership for AI remains untested in the Malaysian courts, the existing legal framework suggests that AI-created works may not qualify for copyright protection in Malaysia as the creator (ie, the AI) does not fall under the definition of “qualified person” under the Copyright Act. 

Another key consideration on the copyright aspect is whether an AI-generated output, once edited and sufficiently transformed by a user, could qualify as an independent creation eligible for copyright protection. Under Section 7 of the Copyright Act, a work may be protected if (a) sufficient effort has been expended to make it original in character and (b) it has been reduced to material form. This raises the question of whether human modification of AI-generated content – through substantial edits, creative input, or reworking – could satisfy these requirements and be recognised as an original work under Malaysian copyright law.

AI and patent law

Similarly, in the field of patents, global discourse continues over whether AI systems should be recognised as inventors. While some argue that AI-created inventions should be patentable, most patent laws – including Malaysia’s Patents Act 1983 – require that an inventor be a natural person reflecting the underlying principle that patent systems are designed to incentivise human creativity and innovation, rather than machine-generated discoveries. This stance has been reinforced by decisions from patent offices in the United States, the European Union and Australia, where AI-generated inventions have been denied patent protection on the basis that only humans can hold patent rights.

Beyond authorship and ownership, AI also introduces challenges in IP enforcement. The rise of AI-powered content generation and deepfake technology makes it increasingly difficult to track and prevent unauthorised reproductions, imitations and counterfeits. For example, AI can now replicate an artist’s style, compose music mimicking a well-known artist, or generate synthetic media indistinguishable from human-created works, raising concerns about how IP laws can be adapted to regulate AI-driven creativity and prevent misuse.

Regulatory considerations and the way forward

Recognising these emerging challenges, below are key considerations Malaysia may take into account in reviewing its IP framework to address AI’s growing impact.

• Should AI-generated works be attributed to the developer, user or AI system itself?
• How should Malaysia balance innovation and copyright protection, particularly for companies training AI models on publicly available data?
• Introducing labelling requirements to differentiate AI-created works from human-created ones.
• Adopting AI-powered detection tools to identify and prevent unauthorised content reproduction and counterfeiting.

Given the global nature of AI and IP challenges, Malaysia may look to international precedents and regional collaborations in shaping its policy response. Future amendments to copyright and patent laws will be crucial in ensuring that IP protection remains relevant in the age of AI-driven innovation.

AI and Ethics

While AI is driving technological progress across industries, it also presents complex ethical challenges that must be carefully managed. Key concerns include bias in AI decision-making, transparency, accountability and the potential for unintended consequences when AI systems operate with minimal human oversight.

A major ethical concern is algorithmic bias, where AI systems may inadvertently produce unfair or discriminatory outcomes due to biases in their training data. For instance, AI-driven hiring tools have been found to disadvantage certain demographics, while automated credit-scoring systems risk unfairly restricting financial access if not properly calibrated. Without rigorous oversight, these biases could reinforce existing inequalities, particularly in sectors such as finance, employment and law enforcement.

Transparency is another critical issue, particularly with AI systems that rely on deep learning models, which can be difficult to interpret. The “black box” nature of such systems raises concerns about explainability, especially when AI is used in high-stakes decision-making like loan approvals, medical diagnostics or law enforcement surveillance. When users cannot understand or challenge AI-driven decisions, it weakens trust and accountability.

Another key challenge is accountability and responsibility. When an AI system makes an error or causes harm, who should be held accountable – the developer, the user or the AI itself? Establishing clear legal and ethical responsibilities is crucial to ensuring fairness and recourse for affected individuals.

To address these concerns, the Guidelines’ Seven Principles and their overarching theme of responsible use of AI is an attempt at advancing human progress without compromising ethics, fairness or accountability. These principles provide a crucial framework for guiding businesses, governments and developers in deploying AI in a way that prioritises transparency, inclusivity and ethical decision-making.

As Malaysia continues integrating AI into public services and corporate decision-making, businesses and regulators must ensure ethical safeguards are in place. Striking the right balance between AI innovation and ethical responsibility will be critical in building public trust and ensuring AI contributes positively to society.

Conclusion

AI is transforming Malaysia’s economy, industries and governance, driving innovation in healthcare, finance and smart cities. The government has demonstrated a strong commitment towards developing AI use throughout the nation through substantial AI-related investments, the AI Guidelines and the establishment of NAIO. While Malaysia has made strides in AI governance, there is an urgent need to establish comprehensive regulatory frameworks to mitigate any potential harms caused by AI. Looking ahead, we can anticipate the rollout of NAIO’s key initiatives, including the AI Technology Plan 2026–2030, which will play a pivotal role in chartering the nation’s AI trajectory in the coming years.