Malaysia to Expand Online Harm Definition; Targets Graphic Content, AI Detection

This move signals a tightening regulatory environment for social media platforms operating in Malaysia, placing increased responsibility on them to proactively moderate content. The focus on local language content detection highlights a specific technical gap that platforms will need to address, potentially requiring investment in more sophisticated, regionally-attuned AI moderation tools. To watch: the specific enforcement actions taken by MCMC against platforms failing to comply and how this impacts content moderation policies globally.

The Malaysian government’s push to expand the definition of online harm and enforce stricter content moderation builds on recent regulatory actions. Effective June 1, 2026, Malaysia's new Risk Mitigation Code (RMC) under the Online Safety Act 2025 mandates online platforms to strengthen measures against scams, harmful content, and manipulated media (Malay Mail, June 2026). The RMC requires platforms to conduct risk assessments, improve reporting and moderation systems, verify advertisers, and label AI-generated content. Non-compliance could result in fines up to RM10 million. Additionally, the Communications Minister Datuk Fahmi Fadzil indicated in January 2026 that the government would review social media licensing requirements, specifically the eight-million-user threshold, to address online harm on platforms with smaller user bases. This was prompted by misuse of AI functions like Grok on platforms like X, which had been exempt from licensing due to reporting fewer than eight million users in Malaysia (Bernama, January 2026). These ongoing efforts underscore Malaysia's commitment to holding social media companies accountable for content disseminated on their platforms.