Malaysia's fight against deepfake content and image manipulation has achieved significant scale this year, with authorities removing more than 12,000 posts in just six months. The Malaysian Communications and Multimedia Commission (MCMC) submitted 13,122 removal requests to social media platforms between January and June, resulting in the successful takedown of 12,353 posts—a 94 per cent success rate that underscores the growing cooperation between regulators and technology companies in combating synthetic media threats.
The volume of deepfake-related requests reveals the mounting challenge posed by accessible manipulation tools and their potential to mislead Malaysian audiences. As synthetic media technology becomes increasingly sophisticated yet easier to use, the ability to rapidly identify and remove harmful content has become essential to protecting public discourse. The high removal rate suggests that social media providers are responding promptly to official requests, though the sheer number of submissions indicates the scale of the problem remains substantial across multiple platforms operating in Malaysia.
Beyond deepfakes, the MCMC's enforcement agenda extends to the broader ecosystem of online deception. During the same six-month period, the regulator submitted 275,787 requests targeting scam-related content, including fraudulent accounts and impersonation schemes. Of these submissions, 262,293 posts were successfully removed—a 95 per cent success rate that mirrors the deepfake takedown performance. This parallel effort highlights how image manipulation and financial scams often operate within the same digital environment, exploiting similar vulnerabilities in how people consume and trust online information.
A significant regulatory development arrived on June 1 with the implementation of the Risk Mitigation Code, which imposes mandatory labelling requirements on licensed platform providers. The code requires all content generated or substantially altered using artificial intelligence to carry clear identification markers, encompassing deepfakes, manipulated imagery, and synthetic audio. For Malaysian users and civil society organisations monitoring misinformation, this transparency measure represents an important step toward making synthetic content immediately recognisable without relying solely on expert analysis or platform content moderation systems.
The labelling requirement addresses a fundamental weakness in the previous enforcement approach: detection and removal occur only after harmful content has already circulated and potentially influenced audiences. By forcing platforms to identify AI-generated material at the point of sharing, Malaysia's regulatory framework now operates at an earlier stage in the content lifecycle. However, the effectiveness of such measures depends heavily on user awareness and platform compliance, areas where Malaysia's media literacy campaigns and corporate accountability mechanisms will face ongoing testing.
Enforcement under Malaysia's legal framework has produced measurable results in false content cases more broadly. Between January 2022 and June 2024, the MCMC investigated 574 cases involving false online content under Section 233 of the Communications and Multimedia Act 1998. Of these, 23 have proceeded to prosecution, with 12 cases concluded and 11 still in trial. The concluded cases resulted in total fines of RM79,000, while one offender received six months' imprisonment after being unable or unwilling to pay the imposed penalty.
The sentencing outcome—jail time for non-payment—underscores the seriousness with which courts have begun treating false content offences. However, the relatively small proportion of investigated cases reaching prosecution indicates the substantial filtering occurring between investigation and court proceedings. As of June 30, authorities had issued 31 compound offers totalling RM1.22 million, deployed 84 warning letters, and maintained 47 cases under active investigation, while many others received determinations that no further action was warranted. This graduated enforcement approach reflects practical constraints and prosecutorial discretion in prioritising cases with the greatest potential harm.
The Online Safety Act 2025, Malaysia's newly operational legislative instrument, provides additional enforcement pathways specifically targeting financial scams. During the first six months of 2024, five removal requests were submitted under this act's provisions, with all five successfully resulting in content takedown. Though the volume appears modest compared to other categories, the 100 per cent removal rate and the act's focus on financial deception suggest it serves as a precision instrument for addressing the most egregious cases rather than a high-volume enforcement mechanism.
Social media companies' cooperation with Malaysian authorities has become increasingly important as regulatory expectations expand. The high removal rates across deepfakes, scams, and false content demonstrate that licensed service providers recognise the legal and reputational stakes involved in non-compliance. However, this reliance on company responsiveness creates vulnerability to changes in corporate policy or investment in content moderation infrastructure. Regional technology companies and Malaysian-based digital firms will likely face pressure to develop local expertise in content assessment rather than relying solely on outsourced moderation networks.
Parliamentary scrutiny has driven the transparency around these enforcement efforts. Senator Musoddak Ahmad's question about artificial intelligence misuse prompted disclosure of the comprehensive data on deepfake removal efforts, while Senator Baharuddin Ahmad's inquiry into fake news combating mechanisms elicited the detailed prosecution record. These parliamentary engagements reflect growing legislative attention to digital harms and suggest that demands for accountability will continue shaping regulatory priorities and resource allocation within the MCMC.
For Malaysian civil society and media organisations, the disclosed statistics provide a baseline against which to assess whether regulatory efforts are adequate to the threat level. The removal of 12,353 deepfake posts in six months, while substantial, leaves open questions about the volume of undetected synthetic content and the effectiveness of detection systems in catching sophisticated manipulations before they spread widely. Public awareness campaigns explaining the existence of labelling requirements and teaching audiences to recognise synthetic content markers will likely become essential complements to regulatory enforcement.
The challenges ahead extend beyond technology companies and enforcement agencies to encompass broader ecosystem factors. Digital literacy remains unevenly distributed across Malaysia's diverse population, meaning that even properly labelled synthetic content may persuade audiences lacking familiarity with manipulation techniques. Additionally, deepfake technology continues advancing faster than detection capabilities in many cases, suggesting that reactive removal approaches must be supplemented by proactive measures such as early warning systems and coordinated research into emerging synthetic media threats.
