As digital technologies become more embedded in daily life, online harms have grown in both scale and complexity. Governments are therefore reassessing how legal and institutional frameworks can better protect individuals.

In recent months, Singapore has introduced legislative changes to strengthen its response. The Online Safety (Relief and Accountability) Act establishes the Online Safety Commission to receive user reports and coordinate action across platforms and agencies, while enabling survivors  to seek relief, including compensation, and to identify anonymous perpetrators for enforcement and legal action.

These measures are complemented by amendments under the Criminal Law (Miscellaneous Amendments) Bill, which clarify that offences relating to the non-consensual dissemination of intimate images apply equally to AI-generated content and expand the definition of child abuse material to include computer-generated depictions.

Taken together, such regulatory changes expand avenues for redress and strengthen accountability. Their impact, however, will depend on whether individuals experiencing harm are able and willing to seek help in practice.

A Significant Proportion of Cases Go Unreported

Consider an example: a man was extorted with an AI-generated sexual video of him. When the deepfake was released, he received a stream of disparaging direct messages on Instagram about his appearance. Members of his social circles had seen the video, and he became fearful of being recognised in public.

At this survivor’s lowest point, he contemplated ending his life.

He did not report the abuse to Instagram or to law enforcement. At the time, navigating reporting processes while in distress felt overwhelming—a situation that is, unfortunately, not uncommon.

SG Her Empowerment’s (SHE) 2023 Study on Online Harms in Singapore found that only half of the survivors reported their experiences to the online platforms where their incident occurred, or relevant authorities such as the police or schools.

This raises critical questions: What factors shape survivors’ decisions not to report, and how can these insights inform more accessible help-seeking pathways?

When individuals experience fear, shame, or distress, tasks that may seem manageable in principle can become significantly more difficult in practice.

A behavioural science lens helps to explain this gap. The EAST framework (Easy, Attractive, Social, Timely) provides a structured way to examine how these conditions shape reporting decisions and why help-seeking often breaks down.

EASY: Current processes feel overwhelming.

A quarter of survivors in the 2023 SHE survey who chose not to report to online platforms cited the process as “too much effort.” Despite efforts by platforms and authorities to streamline reporting, simple steps can feel overwhelming immediately after an online harm incident.

For example, a pre-university student was doxxed in a Telegram group chat. When attempting to apply for a Protection Order under the Protection from Harassment Act, language barriers made it difficult to follow the application instructions.

Behavioural science shows that people are more likely to act when next steps feel simple and immediate. Under stress, a survivor’s cognitive bandwidth shrinks. Lengthy forms, confusing links, or repeated documentation of painful details can stop survivors in their tracks.

ATTRACTIVE: Survivors often see reporting as a futile exercise.

Findings from SHE’s recent study, 404 Help Not Found: Lived Experiences of Online Harms Survivors, highlight a recurring perception that reporting efforts often do not lead to meaningful outcomes.

Some described receiving generic automated responses from online platforms they reported harmful content to, or no response at all, while others were discouraged by the prospect of prolonged legal processes. Survey data shows that among those who reported incidents, only one in four were satisfied with platform responses, and only one in three found legal options adequate.

It is therefore unsurprising that many individuals assess reporting as unlikely to justify the effort involved and choose not to proceed. Behavioural science describes this as ambiguity bias, where uncertainty about outcomes reduces the likelihood of action.

During periods of vulnerability, clearer signals of progress and resolution can make a significant difference. Transparent updates from platforms receiving user reports, timely communication from authorities, and referrals to additional support services can help position reporting as a constructive step rather than an endpoint with uncertain value.

SOCIAL: Reactions matter, especially from close ones.

The top reason for not reporting, according to SHE’s 2023 survey, was survivors’ belief that the incident they faced was “not serious enough.” A prevailing sense of acceptance or “normalcy” surrounding online abuse leads to a wider feeling of resignation, which, in turn, discourages survivors from seeking help.

When minimisation or dismissal of abuse is the norm, survivors hesitate to speak up.

For example, a survivor coerced into sharing intimate images hesitated to report, worried that others might blame him/her or dismiss their experience. Conversely, encouragement from a friend or family member often prompted survivors to seek help, sometimes leading them to support centres such as SHECARES.

An approachable, friendly presence can mean a lot for survivors. Validating a survivor’s experience, encouraging reporting, or even flagging harmful content when we see it are steps we can all take to show support.

TIMELY: A swift response to harm is needed.

Online harm escalates rapidly. Survivors often describe a loss of control and agency, especially in the first hours after an incident. In this disorienting state, the prospect of reporting often feels overwhelming.

Behavioural science explains this pattern through present bias. The immediate stress, uncertainty and the emotional strain of filing a report often outweigh the potential relief that could come later.

Timely content removal helps counter this bias. It keeps pace with how quickly online harms unfold, bringing the benefits of action into the present before inertia takes hold.

Why these Insights Matter for Policy

These behavioural barriers help explain why online harm often goes unreported, even when its impact is severe. Understanding the factors that influence survivors’ decision to seek help is essential before evaluating whether current and emerging policy responses meaningfully meet their needs—or risk compounding the challenges in practice.

By factoring in these real-world experiences, policymakers can design interventions that are not just well-intentioned, but effective—removing friction, providing clear guidance, and ensuring survivors feel supported at every step.

Read more in the second part of this Reporting Online Harms series, where we examine how current approaches to online safety respond to these lived realities, and what this means for enabling effective help-seeking.

The views and recommendations expressed in this article published on are solely of the authors and do not necessarily reflect the views and position of the Tech for Good Institute.