Welcome Address by Minister for Law and Second Minister for Home Affairs Mr Edwin Tong...

23 September 2026 Posted in Speeches

Introduction

1. Thank you very much, everyone. It is such a pleasure to see all of you here, present together in the same room. Many of you worked with us, and your input was so valuable as we constructed the Online Safety (Relief and Accountability) Act (OSRAA) from scratch over several years.

2. We organised today's event for a very simple reason. It is to express our appreciation and to say a big thank you to all of you for working with us, sharing your stories with us, being so generous with your time and your expertise, as you helped us front-end as well as back-end to put this piece of legislation together.

3. Many people in this room came together to help shape OSRAA - studying the problem from start to end, consulting the stakeholders, understanding what the harms meant and the impact it made, working through some very difficult policy and legal questions, drafting the legislation, building the systems and the processes, training the teams, and then, of course, preparing the Online Safety Commission (OSC) for its very first case.

4. As you have seen from some of the numbers published, it is off to a flying start. Since the OSC began operating almost three months ago, many of you have been putting the work that we have done into practice. We spent such a long time in gestation, thinking about it, and preparing it that so much has changed. In fact, so much has changed that some of my colleagues who were working at the Ministry of Law (MinLaw) are no longer my colleagues and have become part of OSC like Francis (Francis Ng, Online Safety Commissioner). We lost Francis - our loss, but OSC's gain. So congrats, Francis and the team at OSC.

5. Indeed, today is about recognising all of you for playing a big part in that effort - our community as well as social service partners; the platforms; academia; our colleagues at the OSC, MinLaw and the rest of Government; and the many other stakeholders who told us your stories and gave us your suggestions.

What should success look like

6. Yesterday, we shared publicly some of what the OSC has seen in its first two months.

7. I will not go through the figures again, but I thought it is important for me to highlight three areas which are relevant to what I will speak about today. First, more than 500 reports were received in the first two months. Second, more than 90% of eligible cases were resolved. And third, more than 80% of eligible reports involved doxxing, in some cases, very young victims, or online harassment.

8. Now, each of these numbers tells us something. The first tells us that there was indeed a need for this avenue of relief, and that people knew about the OSC and are prepared to come forward. I think that is very encouraging as we work together hand in hand to keep the online space safe. The second gives us confidence that we are on the right path, and it is kudos to the OSC team that the system is delivering relief. But the third begins to tell us something about the nature and patterns of the harms that we are seeing.

9. Now, these are still very early numbers. Beyond numbers, the more important question is what we do with what these numbers are telling us.

10. I will also be careful about treating a high number of reports, by itself, as a measure of success. It is not that we have 500 cases, and if we can handle more cases, it is successful. In fact, our longer-term ambition should be the very opposite. We should drive the number of cases down. We should want people to rely on or need the OSC less and less, not more and more. The OSC is, of course, there when people need help. But ideally, the community, working with the platforms, should be able to deal effectively with harmful content before a victim needs to escalate the matter to the OSC. Even further upstream, we should seek to reduce the occurrence of the harm in the first place.

11. What are some of the societal norms that we can shape? Can we encourage a set of behaviour and norms? What is acceptable behaviour online and can we discourage bad actors through deterrence? We want people to understand that what you do behind a screen will have consequences and anonymity does not mean impunity.

12. That is where we want to get to: cutting down the numbers, creating a platform that becomes, over time, self-regulating, and people understanding that this is the kind of social behaviour that we should encourage.

13. There are three areas that I thought I will speak about today.

a) The first is how can we improve the relief? I think we have had a good start, but we've always asked ourselves, as we were building this and spending so much time in the early parts of the consultation to think about the relief: "How do we enhance the relief to our victims?" I think that is perhaps the most important question.

b) Second, allied to that, is what can we learn from the cases that we have seen so far? 500 cases that gave us material to learn from, so that we can advance when the next 500 comes.

c) Third, how can we use those lessons to prevent more harm so that the 500 cases today do not become 500 cases tomorrow? Perhaps through the lessons we learn, we can reduce the number of cases.

We will make relief faster and easier

14. Let me start with relief.

15. As I said, the early resolution rate has been very encouraging. I hope you see that too. But our experience over these first few months also tells us that there is room to make the process faster and easier for victims. I think that is a very important measure of our success - what it means to the victims, how they navigate the system, and how quickly and effectively they can get relief.

16. So let me touch on the first area, which is speed.

17. Today, prescribed platforms must respond to Online Harms Notices within 6 hours for straightforward cases of intimate image abuse and image-based child abuse; 24 hours for other cases involving these harms; and 48 hours for harassment, doxxing and stalking.

18. We calibrated these timelines very carefully. We knew always, from the start, from the victims' accounts, that speed mattered a lot. But we also recognised that the platforms needed time to understand and to come to terms with the new framework and adapt their processes so that they will be able to function within this framework. So these timelines that we had introduced are a starting point.

19. As platforms become more familiar with the framework, we will work with them towards shorter response times. Because when harmful content is circulating online, even that extra one hour can matter, and sometimes can matter a lot.

20. We also want to make it easier for victims to give notice of harms.

21. The prescribed Online Harm Notice is available on MinLaw's website, and we are open to platforms hosting webforms on their own services.

22. But wherever the form sits, the principle should be that it is simple: make it easy to find, easy to use, and ask for only information which is reasonably necessary. Do not ask for a whole list of information which perhaps you do not need at the start or at all.

23. For the major platforms, we also hope that they can provide a clear, dedicated channel - perhaps maybe a designated email address for a start - for people to submit these Notices easily. So, the moment you need to reach a particular platform, you find the email address, and after confirming it is the right channel, you can put your Notice through.

24. We should approach Notices in a practical and most importantly, victim-centric way. We have encountered cases where victims perhaps can no longer locate the original content. Sometimes it goes around, you cannot find it anymore or it is blocked. But you know screenshots still travel and sometimes they travel very virulently. Maybe privacy settings may have since changed because of the attention. But all of these hurdles should not put relief beyond their reach. Just because you cannot find the original post or access it does not and should not stop you from seeking assistance.

25. Where there is a credible report, the focus should be on whether there is enough information to investigate and to act, and not whether every procedural box has been checked. We need not expect every case to always have all the procedural boxes checked before action is taken.

26. Another practical barrier involves respondents who are overseas. In today's world, online harm does not respect borders; it crosses borders very easily. But serving court documents across borders can sometimes be much harder.

27. So we are looking at greater scope for electronic service out of jurisdiction, so that legitimate claims are not frustrated by such impediments, while still ensuring that the documents are effectively brought to the respondent's attention - that means proper notice is given and that it is fair.

28. The principle is straightforward: being overseas should not, by itself, put someone outside of accountability.

29. These may sound like procedural changes. But to a victim, it eases the way and removes one hurdle to surmount. It can make a real difference because they give people the ability to act meaningfully when they suffer online harm.

30. That lies at the heart of OSRAA's novel design. The OSC provides a route to timely relief but alongside it, there is a clear statutory tort framework that empowers victims to take action themselves against those responsible for the harm.

We will learn from every case

31. Second, let me touch a bit about what we will learn from the cases that we have seen before us today. Because this is a new framework, and a very novel one at that, we should also be prepared to learn from these cases, even at a very early stage, and improve the framework bit by bit as we go along.

32. The first numbers from the OSC tell us something. But the real value is not in the numbers themselves; it is in what these cases can teach us.

33. Because each time a case reaches the OSC, we should ask ourselves, "Why did it have to reach the OSC? Was the harm difficult to report or not recognised by the platform? Did the response take too long? And after action was taken, did the perpetrator simply reappear through another account or repost it on some other platform?"

34. Every escalation is not just a case to resolve, but it is also a gap that we can understand.

35. When we look across cases, rather than each case in isolation, we can also begin to study the patterns that emerge from these 500 cases.

36. Ultimately, we do not want to play whack-a-mole with the bad actors and with harmful content. When you remove one post, it turns up in another place. Or you remove one account and another is created. Or when you remove material from one platform, and you find it again on some other platform. We need to study how we can achieve a more holistic solutioning to the problems that victims face and give them a more effective remedy.

37. Of course, removing the harmful content matters enormously to the person affected. But if we stop there, we may solve the immediate problem without understanding what is driving it. If you take them individually, there are 500 cases to resolve. Taken together, they can tell us a story, and that is what I meant by looking at the pattern. The concentration of reports in particular types of harm is one example of why we need to look across cases.

a) Are we seeing repeat perpetrators, for example, or harms that simply move across accounts on different platforms?

b) Is technology, including AI or specific platform features, changing the nature and the scale of the harm?

c) Where are the gaps in the system, and what can we learn from platforms that are responding particularly well?

d) How can we come together as a community to learn from one another and to benefit from the best practices of each other?

38. That is why, when we look back at these first 500 cases, I am interested not just in another set of numbers or statistics. I would like to know what surprised us.

a) What did we not anticipate after having spent such a long time doing the consultations? What is new about it? What is surprising about it?

b) What has today's technology allowed perpetrators to do, which we did not anticipate when we were doing the consultations?

c) How are the harms, the perpetrators and the technology evolving in ways we did not expect?

d) And just as importantly, what have we seen that works particularly well and helps victims, and what can we do better?

39. We do not have all the answers yet - we are just three months in after all. But as that picture becomes clearer, we will share what we are seeing from the patterns, from the cases, from the trends. Not the details of individual cases, of course, but the broader patterns and insights that can help everyone do better. We will share them with platforms, because that helps us to mutually strengthen our systems to address recurring gaps. We will share them with our community partners, so that they can better support people who experience online harm. And we will share them with the public, so that people can understand the risks, and see the evolving nature of online harms - how they are perpetrated, and what they can do if they encounter such harms.

40. If the same type of case keeps reaching the OSC for the same reason, we should not simply become better at processing it. It is not a numbers game. It is not better if we only resolve it faster. Instead, we should try to close the gap that keeps producing those cases.

41. If we do that well, the first 500 reports that we have seen so far will not simply be 500 cases that we have dealt with and closed. They will help us deal better with the next 500, and perhaps prevent some of the next 500 from even occurring.

42. Perhaps that is ultimately how we should measure success: not by how many people come to the OSC for assistance, but by whether, over time, fewer people need to do so.

Conclusion

43. I mentioned earlier that this piece of legislation was novel, unique, and progressive. But none of that could have happened simply because we passed a law or established an OSC.

44. It happened because of the numerous people who have stepped forward, in some cases, out of their comfort zone, to come and help us.

45. I would like to conclude this afternoon by thanking each of them in turn.

46. First, and I think very importantly, I want to acknowledge the victims, who stepped forward and shared their experiences with us. It takes immense courage to even talk about something that has been so deeply distressing and highly personal. But many did so, knowing that, by sharing the experiences, it probably will not take away their harm, but it will help someone else in the future navigate their problems better. We are very grateful for that trust you put in us when you shared those stories. Your stories and experiences give us even greater conviction that we are doing something that is right, and that we have to get this right.

47. Second, I want to thank YouthTechSG, and all our youths who have contributed. Thank you very much for speaking candidly about your experiences, and how you navigate the online world. Having that perspective was very useful. It really gave us a different insight into the realities on the ground.

48. I want to also thank our colleagues in academia - SMU Yong Pung How School of Law, Ms Simran Toor, Professor Lim Sun Sun, Dr Chew Han Ei and Dr Carol Soon and others. Thank you for bringing the intellectual rigour and your experience to the work that we were doing, contributing your research and your comparative perspectives, and in many cases, challenging our assumptions. It has been very constructive. As I said, we have a piece of legislation that is both progressive and novel. OSRAA was, by design, not only intended to give relief, which the OSC can do, but what I like about it too is that it empowers victims. It puts in their hands the ability to invoke the statutory tort framework for self-remedy, and when you put that power in the hands of victims, it sends a signal as well. It sends a message that they are not helpless, and that they can do something about the harms that they have experienced.

49. To our many social service agencies and community partners - SHE, TOUCH Community Services, Fei Yue Community Services, The Singapore Children's Society, and The Samaritans of Singapore, and many more - often, you are the trusted first responders. You are the interface between the remedies that we can make available or the assistance that you can find and the traumatic experience for the victim. You are that interface. Thank you very much for sharing your experiences with us, and for being the bridge between the avenues of relief we have created and the people who need them.

50. To the platforms - Meta, Google, Tencent, Hardware Zone and many more - thank you for working with us collaboratively. I think the responsibility that we both carry is a very heavy one, and that is to make the online space safe. It is a tremendous responsibility but one that we are grateful that we are able to share with the platforms, and that they are of like mind that this is what we need to do to keep our online space safe. Thank you to all of the platforms for doing that.

51. Finally, to our colleagues at the OSC, MinLaw and across all of Government, we worked really hard, and really kudos to all of you. There were many late nights. We consulted and reconsulted, drafted and redrafted. We tested this and tested that. I think I changed my mind on a few things quite late in the day, close to the First Reading in Parliament, so sorry about that. But it is all because we were trying to respond to the various constituencies of victims or potential victims that are out there. As we listened to them, we were trying as much as we can to cover the gaps. I know it will not yet be complete, and we will always have room to improve. But we tried very hard to ensure that we started with something that is workable and provides us a strong framework, as we experiment with the initial categories of harm and look at expanding the coverage later on. There has been tremendous amount of thought, effort and deliberate care taken in the drafting and the nuancing, so thank you to all my colleagues in Government for doing that.

52. There are many others who have contributed. I cannot name you individually today, but please do know that your contribution is seen, and deeply appreciated. To all those who I might not have named, today's occasion is really for us to thank you in person.

53. Finally, there is one group that I have not thanked, and it is not often that they get thanked at events like this - our friends from the media. By reporting on these issues and sharing, in a very conscious and very careful way, the experiences of those affected, you help more people understand the nature of the harms, and how these harms can affect them when they happen online. Just as importantly, you help people know that help is available, and where they can turn to when they need it. Very often, you do not think about it until you actually need it. And when you need it, it becomes the most important thing in the world. So it is important that we have a media that helps us to propagate that information. That awareness makes a real difference. So thank you to the media, for helping us to shine a light on these cases.

54. I am going to end by saying that today really is truly an appreciation event. I hope all of you will join us for a short tea reception later on. I hope to catch up with you and maybe share some more stories with you later. But once again, thank you very much from us in Government to all of you for working with us. When you look around the room, you do not just see the sheer numbers, but also the spectrum of society that all of you represent. That really was the spectrum that went into creating a Bill like this. Today, it is in practice, and it is as much your product as it is ours.

55. On that note, thank you very much. Enjoy the evening.

Last updated on 23 September 2026