Malaysia moves to regulate AI with proposed governance bill

Malaysia is reportedly set to introduce a legal framework to regulate artificial intelligence (AI), with proposed legislation covering AI systems, their lifecycle, as well as the developers and deployers responsible for them.

The proposed Artificial Intelligence Governance Bill, outlined in a public consultation document released by the National AI Office (NAIO), aims to establish a national framework for the safe, responsible and innovation-enabling use of AI.

According to The Star, the legislation will apply to AI systems placed on the Malaysian market or put into service in the country, as well as systems designed, developed or used in Malaysia. It will also cover AI deployed by organisations established in Malaysia, regardless of where the systems are hosted. The bill will not apply to the personal use of AI or matters relating to national security.

Don't miss: Using AI in your content? You could be dampening brand trust

The proposed legislation is reportedly built around five governance principles: protecting human dignity and rights, transparency and explainability, accountability, safety and security, and responsible data governance. Together, these principles are intended to support trustworthy AI development while fostering innovation and public confidence.

Rather than prescribing detailed technical rules, the bill adopts a principle-based approach, with implementation supported through standards, guidelines and other regulatory instruments.

It also reportedly introduces a risk-based regulatory framework, under which AI systems will be classified as presenting unacceptable, high or low risk. Regulatory obligations will increase in line with an application's risk level, meaning higher-risk AI systems will face stricter governance requirements while lower-risk applications will be subject to lighter obligations.

The risk framework is anchored around the potential for harm, including death, bodily injury, unlawful deprivation of fundamental liberty under Malaysia's Federal Constitution, and breaches of existing laws.

In addition, the bill reportedly proposes mandatory reporting of AI incidents, including failures, weaknesses, misuse, unexpected effects and near misses. Organisations would be required to disclose details such as the nature of the incident, foreseeable harm, containment measures, root causes where applicable, and remediation actions taken.

The framework also proposes establishing a central AI authority to oversee national AI governance. According to The Star, the authority would be responsible for maintaining the country's AI risk framework, supervising assessments, supporting testing, developing incident reporting mechanisms and strengthening international technical cooperation.

The authority would also reportedly oversee investigations and enforcement, including technical fact-finding following AI-related incidents, while providing guidance, training and compliance support for regulators, public agencies, AI sandbox operators and organisations subject to the framework.

To support sector-specific implementation, the authority may appoint "Sectoral leads" with delegated powers where they possess the necessary legal authority, technical expertise and governance capabilities. According to the consultation document, the proposals remain subject to further review and amendment before becoming law.

The framework is reportedly intended to operate alongside existing sectoral regulators, allowing public agencies to coordinate AI governance under a common national framework while retaining the flexibility to address industry-specific risks and operational needs.

The proposed bill follows earlier indications that Malaysia was preparing dedicated legislation to regulate AI. Last week, communications minister Fahmi Fadzil said the government was working on a dedicated artificial intelligence bill to establish a regulatory framework addressing the growing risks associated with the technology.

The proposed legislation, which is being drafted by the Digital Ministry, aims to address concerns ranging from online scams and harmful content to the misuse of personal data as AI capabilities continue to evolve faster than existing regulations.

Related articles: 
Can AI avatars bring politicians and citizens closer? Malaysian PM Anwar thinks so   
Malaysia reportedly eyes AI copyright safeguards under proposed legal reforms 
X pledges safeguards in response to Grok safety concerns, says Fahmi