In November 2024, parents at Singapore Sports School received a call no parent should ever get. Someone had used deepfake technology to create nude images of their daughters—and the fake photos were circulating among classmates.
The police launched an investigation; the school launched an internal inquiry; and across the city, a quiet question began to take hold: if a school campus wasn't a safe space from AI-generated abuse, where was?
A year later, a survey by the Ministry of Digital Development and Information (MDDI) delivered an equally sobering data point: only 37 percent of parents felt confident managing their children's digital activities—and more than half wanted stronger government support.
The deepfake case was a symptom. The survey was a measure of public anxiety. Together, they help explain why Singapore has, in just a few years, built one of the world's most layered responses to online harm.
The island's approach rests on four pillars: legislation, public education, community partnerships and regulatory oversight. However, the real story is not the number of pillars. It is how quickly the architecture has evolved—and how deliberately it is being anchored, not in censorship, but in accountability.
Harms that led to the upgrade of the old law
Before 2022, Singapore's Broadcasting Act did not apply to online communication services operating overseas. While the Broadcasting Act could be used to block websites from being accessed by Singapore users, it did not allow for action to be taken against specific harmful content on these services.
The Online Safety (Miscellaneous Amendments) Act, passed in November 2022 and effective from February 1, 2023, closed that gap. For the first time, Singapore's Infocomm Media Development Authority (IMDA) could issue directions to disable access to "egregious content"—which is defined under the Broadcasting Act as material advocating suicide or self-harm, depicting child sexual exploitation, inciting racial or religious disharmony and posing public health risks, among others.
The law applied not just to platforms but to individual posts, accounts and even groups. Crucially, IMDA could now require overseas-hosted online services accessible to Singapore users to comply with Codes of Practice, which require such services to put in place systems and processes to mitigate exposure to harmful content and undergo audits and risk assessments.
However, legislation without enforcement is theater. In March 2026, IMDA issued formal letters of caution to X and TikTok for serious weaknesses in their detection of child sexual exploitation material (CSEM) and terrorism content, respectively.
IMDA found a 120 percent increase in CSEM cases on X originating from or targeting Singapore users—from 33 cases in 2024 to 73 cases in 2025. On TikTok, 17 cases of terrorism content shared by Singapore-based accounts were detected for the first time. Both platforms were placed under "Enhanced Supervision" and required to report their progress until IMDA was satisfied.
A second law, written for victims
The 2022 Act was about removal. The 2025 Act—the Online Safety (Relief and Accountability) Act (OSRA), passed in Parliament on November 5, 2025—is about redress.
The law establishes the office of the Commissioner of Online Safety, which is supported by a new agency, the Online Safety Commission (OSC), which began operations on June 29, 2026. It also introduces statutory torts, which allow victims to sue perpetrators in court for harms caused to them.
OSRA covers 13 categories of online harms, ranging from online harassment and doxxing to intimate image abuse and image-based child abuse. The definition of intimate image abuse explicitly includes AI-generated imagery—the very technology used in the sports school deepfake case.
In the first phase of operations, the OSC will focus on five of the most prevalent and serious online harms. Following an assessment of a victim's report, the OSC may direct that access to the harmful content be disabled or that the perpetrator's account be restricted. Where appropriate, and if requested by victims, the OSC may direct platforms to provide information to identify perpetrators of online harms.
From regulation to prevention
Regulation alone does not build a safer digital society. Singapore has also invested heavily in the softer infrastructure of online safety: education, community outreach and technical tools.
In May 2026, Prime Minister Lawrence Wong launched "Screen Smart from the Start," a nationwide movement that aims to help parents and children build healthy digital habits together. This movement is also accompanied by a new one-stop portal, BeScreenSmart.gov.sg, offering age-specific guidance for children aged 0-6, 7-12 and 13-18.
The movement responds directly to the survey finding that more than half of parents wanted more help, and that only 37 percent felt confident. For schools, cyber wellness has been embedded in the mandatory Character and Citizenship Education curriculum for years, while the Media Literacy Council has developed resources under the Digital for Life movement, working with partners including Google and Meta.
On the technical front, Singapore has built the Center for Advanced Technologies in Online Safety (CATOS), a government research hub housed at A*STAR and funded with 50 million Singapore dollars ($39 million) over five years under the Smart Nation and Digital Economy domain of Singapore's Research, Innovation and Enterprise 2025 Plan. CATOS focuses on detection and prevention tools for hate speech, deepfakes, and disinformation.
There are also citizen-led initiatives like CheckMate, a WhatsApp-based fact-checking service where users can submit suspicious messages, images or videos to verify their authenticity.
However, the most significant change may be at the app store level. From April 1, 2026, five app stores—Apple, Google, Huawei, Samsung and Microsoft—designated under Singapore's Code of Practice for Online Safety: App Distribution Services, are required to prevent under-18 users from downloading age-inappropriate apps. The stores use a range of methods, including Singpass authentication, facial age estimation, government ID uploads and credit card verification.
This is not a content moderation policy. It is a gatekeeping policy—and for a country where more than 90 percent of the population owns a smartphone, it amounts to a quiet revolution in how young people enter the digital world.
No one law, no single fix
What makes Singapore's approach unusual is not any single intervention but the coherence of the cascade. The legal floor is set by the 2022 Act, which allows the rapid removal of the worst material. The 2025 Act gives victims a low-cost, high-speed path to justice and accountability. The technical infrastructure is being built at CATOS.
The education piece reaches parents through community workshops and schools through mandatory curricula. In addition, the app store age assurance measures are a preemptive strike—blocking harmful exposure before the first click.
The sports school deepfake case is still winding its way through the system. Nevertheless, with the OSC opening its doors on June 29, 2026, victims will have a dedicated agency they did not have before. The survey parents who felt alone will have a portal, a movement and a growing set of tools. The platforms that failed to detect CSEM and terrorism content will remain under enhanced supervision until they prove they can do better.
Singapore does not claim to have solved online harm, and no country has, but in its methodical, multi-layered way, the city-state is trying to answer a question that every digital society will eventually have to face: how do you protect the most vulnerable in a space that was designed, from the start, to have no rules at all? The answer, it turns out, is not one law. It is a system—and a willingness to keep rebuilding it.
Reporter | Guo Zedong
Photo | Guo Zedong
