AI Governance Bill to introduce risk-based regulation
The proposed legislation is scheduled for tabling in the Dewan Rakyat and would complement existing sectoral laws.
Malaysia is preparing legislation covering AI safety, individual welfare and human rights, with regulatory requirements tied to the harm a system could cause, Digital Minister Gobind Singh Deo said. The Artificial Intelligence (AI) Governance Bill is scheduled for tabling in the Dewan Rakyat.
Gobind told the Dewan Rakyat that the framework would span all sectors while working alongside their existing laws. Controls and obligations would reflect the potential harm from each AI system, rather than relying only on an industry's risk classification.
He said wider use of more powerful or autonomous AI would require stronger safeguards, especially where people or critical systems could be significantly affected. These would involve testing and monitoring, oversight by humans and arrangements to respond to problems.
Gobind also said the Ministry of Digital and AI Malaysia Berhad had established the Malaysia AI Safety Institute. The institute would examine mechanisms for regulating AI models used in organisational operations across sectors, with responsibilities covering safety benchmarks, assessment of risks and incident reports to support corrective action.
His remarks answered Datuk Dr Ku Abd Rahman Ku Ismail (PN-Kubang Pasu), who asked about calls to slow AI development while regulation, ethical safeguards and plans for worker protection are prepared.
