The Malaysian Communications and Multimedia Commission (MCMC) has ramped up its enforcement against deepfake content and online fraud, successfully removing 12,353 manipulated posts during the first half of 2024. According to a parliamentary reply from the Ministry of Communications released this week, these takedowns represent a 94 per cent success rate following 13,122 removal requests submitted to social media platform providers between January 1 and June 30. The aggressive pursuit of deepfakes reflects growing concerns about the weaponisation of artificial intelligence in spreading false information and impersonating individuals across Malaysian digital spaces.
The scale of the crackdown extends well beyond deepfakes alone. During the same six-month period, the MCMC submitted 275,787 separate requests for removal of scam-related material, including fake accounts and impersonation content designed to defraud users. The commission achieved a similarly impressive removal rate of 95 per cent, taking down 262,293 posts. This dual offensive against both image manipulation and financial fraud indicates that Malaysian authorities view AI-enabled deception as part of a broader ecosystem of online harm that requires coordinated, multi-pronged enforcement. The statistics suggest that platform providers are cooperating relatively effectively with government requests, though the sheer volume of content requiring intervention underscores the challenge of keeping pace with bad actors who continuously generate new fraudulent material.
The enforcement effort has been strengthened by new regulatory requirements introduced this year. The Risk Mitigation Code, which took effect on June 1, mandates that licensed platform providers label all content generated or altered using artificial intelligence technology. This includes deepfakes, manipulated images, and doctored audio files. The labelling requirement aims to alert users to potentially unreliable content and create a friction point that may discourage the sharing of synthetic media. For Malaysian users, the practical effect should be greater transparency about which posts have been artificially created or significantly altered, providing essential context as AI tools become increasingly sophisticated and harder to detect through visual inspection alone.
Parallel to these content removal efforts, authorities have invoked more recent legislative tools to target specific categories of harm. The Online Safety Act 2025, which represents Malaysia's latest attempt to regulate digital spaces, has been deployed selectively against financial scams. Between January and June this year, five removal requests were filed under this legislation, with all targeted content successfully taken down. While the number of Online Safety Act interventions remains small compared to broader removal efforts, the statute provides a more tailored legal framework specifically designed to address emerging online safety challenges and gives regulators additional enforcement mechanisms beyond the Communications and Multimedia Act 1998.
Beyond content removal, the MCMC has pursued criminal prosecution of individuals responsible for spreading false information online. Between January 2022 and June this year, the commission investigated 574 cases involving false online content under Section 233 of the Communications and Multimedia Act 1998. Of these, 23 cases proceeded to court, with 12 concluded and 11 still pending trial. The courts have imposed financial penalties totalling RM79,000 across the concluded cases, while one offender received a six-month prison sentence after refusing to pay the imposed fine. The mix of fines and imprisonment demonstrates that Malaysian authorities view deliberate online falsehoods as serious enough to warrant custodial sentences, not merely financial penalties.
The enforcement landscape reveals a tiered response system reflecting the seriousness of different categories of offence. In addition to prosecutions, authorities have issued 31 compound offers totalling RM1.22 million, allowing offenders to settle charges through payment rather than court proceedings. Eighty-four warning letters have been issued to users or content creators, serving as early intervention before more serious enforcement steps. Forty-seven cases remain under investigation, suggesting that the MCMC's work is ongoing. A significant number of cases have been classified as requiring no further action, indicating that not all flagged content necessarily violates applicable law or platform standards. This graduated response suggests proportionate enforcement that distinguishes between intentional fraud, reckless spreading of falsehoods, and borderline or protected speech.
The Malaysian response to deepfakes and online fraud must be understood within the broader Southeast Asian context. Across the region, countries are grappling with similar challenges as AI tools become more accessible and election cycles intensify. The sophistication of deepfake technology has advanced rapidly, enabling the creation of convincing synthetic videos of public figures that can influence political discourse. Malaysia's 94 per cent removal success rate compares favourably with international benchmarks, though the absolute volume of content requiring intervention suggests that reactive removal remains fundamentally insufficient without deterrence and prevention mechanisms. For Malaysian citizens and digital economy stakeholders, the implication is that online content authenticity cannot be assumed; verification and critical consumption of media become essential skills in an environment where synthetic content is actively being weaponised.
The case of HarakahDaily's Facebook account illustrates the practical application of these enforcement frameworks. According to the parliamentary reply, as of June 30 this year, no First Information Report had been filed regarding the news outlet's account, but the MCMC indicated that firm action would be taken if any content was identified as breaching applicable law or platform guidelines. This statement suggests that high-profile accounts remain subject to the same scrutiny as others, though the absence of an FIR to date indicates either compliance or that any alleged violations did not meet the threshold for formal police reporting. The emphasis on taking action if content breaches law or platform standards reflects the dual legal and contractual frameworks within which social media regulation operates in Malaysia.
Looking forward, the sustainability of Malaysia's enforcement approach depends on several factors. The 94 per cent removal success rate reflects good cooperation with platform providers, but maintaining this level of responsiveness will require ongoing dialogue and technical integration between regulators and tech companies. The shift toward labelling AI-generated content under the Risk Mitigation Code represents a more preventive approach than purely reactive removal, though the effectiveness of labels in changing user behaviour remains an open question. For Malaysian businesses and content creators operating in digital spaces, the clear message is that inauthentic content will face removal and potential legal consequences. The enforcement statistics demonstrate that Malaysia's regulatory apparatus is active and increasingly sophisticated in its use of both traditional criminal law and newer regulatory codes specifically designed for the digital age.
