Southeast Asia Cannot Fight Online Sexual Abuse One Takedown at a Time

When screenshots of a private group chat involving 16 male law students at Universitas Indonesia (UI) surfaced in April 2026, the content was familiar: sexual jokes about female classmates and lecturers, and a line declaring that silence means consent. The response was not. Women students publicly confronted the men at a faculty forum, the university suspended all 16, and the Indonesian Education Monitoring Network called for a state of emergency on violence in education. A generation ago, such vulgarity might have eluded accountability.

This moment did not arrive in isolation. Across Southeast Asia, governments are beginning to recognize online sexual harassment as a legal harm, but their responses intervene at different stages of the abuse. Thailand criminalised sexual harassment, including conduct through electronic communications, in December 2025. Its “Take it Down” procedure, launched the following month, allows victims to apply online for a court order suspending dissemination and removing offending content without waiting for a police investigation. Malaysia’s Tribunal for Anti-Sexual Harassment offers an alternative non-court avenue for apologies, compensation and other remedies against identifiable perpetrators. Singapore’s Online Safety Commission can direct perpetrators, group administrators and platforms to disable harmful content or restrict accounts.

Indonesia’s Sexual Violence Crime Law (Law No. 12 of 2022) similarly recognises electronic-based sexual violence and grants victims the right to seek the removal of content containing sexual images or descriptions. A 2025 implementing regulation has since clarified the procedure for restricting or deleting such content.

The region’s regulatory systems offer three different routes. Thailand relies on an expedited judicial process, Singapore on a dedicated administrative intermediary, and Indonesia on public requests to state authorities. Each can provide meaningful relief, but the remedies all begin after the harmful material has been created, discovered and reported.

This shared limitation explains why the region’s emerging framework is useful but incomplete. Criminal laws establish culpability, tribunals provide redress and takedown procedures can limit further circulation. None can, by itself, prevent abusive content from being copied, recommended, monetised or recreated. Most mechanisms also remain complaint-driven, leaving victims to find the material, preserve evidence and pursue repeated reports as it spreads across services and jurisdictions.

The real test of regulation is not whether a law exists, but whether harm can be stopped before it becomes permanent.

That burden is especially difficult where criminal justice systems are measures of last resort. Across Asia, victims often do not report abuse due to fear of reprisals, limited confidence in the police and courts, the cost of legal action and victim-blaming attitudes in society. Thailand’s expedited court process is therefore a valuable option, but reproducing a court-centred model throughout Southeast Asia could strain judicial systems and exclude those least able to navigate them. Judicial orders should sit alongside administrative bodies, tribunals, civil remedies and direct platform assistance rather than become the principal gateway to protection.

Generative AI makes the need for these parallel remedies more urgent. In 2024, students at the Singapore Sports School generated and circulated deepfake nudes of female classmates. The problem widened in early 2026, when Indonesia, Malaysia and the Philippines restricted Grok after it was used to generate non-consensual sexualised images of women and children. The restrictions provided a temporary brake, but not a durable solution. Users could circumvent geographic blocks, while reliance on reporting meant intervention began only after images had been generated and circulated.         

The Grok episode exposed the central regulatory gap. Notice-and-takedown rules address individual pieces of content, but do not require platforms to confront the systems that enable abuse to recur. Removing one image does not prevent a generative tool from producing another, nor does it stop recommender systems from amplifying sexualised material or prevent advertising systems from rewarding harmful engagement. A framework centred on individual complaints consequently treats recurring abuse as a succession of isolated incidents rather than a foreseeable act and governance failure.

Platform accountability must therefore extend beyond faster takedowns. Companies should be required to assess gendered risks before deploying products, build safeguards against the creation of sexualised images of real people, preserve evidence, restrict repeat offenders and provide regulators with information needed to evaluate compliance. Duties should also cover recommender systems and digital advertising, particularly where their design exposes women and children to predictable harm. The objective is not to replace victim remedies, but to reduce how often victims need them. Australia’s Safety by Design framework and the European Union’s Digital Services Act show how regulation can shift from post-by-post removal to systemic risk reduction.

Institutions closer to the point of harm must also be able to respond. The Universitas Indonesia and Singapore Sports School cases were assembled within student networks before formal authorities intervened. Schools and universities need procedures for protecting victims, preserving digital evidence, responding to synthetic sexual imagery and imposing sanctions. Police, prosecutors and judges require corresponding training so that new remedies do not reproduce the barriers that survivors already face.

ASEAN cannot create a single regional criminal code, but it can support urgent cross-border evidence requests, shared preservation standards and comparable prevalence data. Southeast Asia does not simply need more offences. It needs connected remedies that victims can access without depending entirely on criminal proceedings, capable institutions that can act quickly and enforceable requirements on platforms to prevent recurrence. The real test of regulation is not whether a law exists, but whether harm can be stopped before it becomes permanent.

2026/211