Online Safety Policies

❓ Is the Online Safety Act working ❓ One of my roles in the UK House of Lords is as a member of the Communications and Digital Select Committee. We are currently scrutinizing the implementation and effectiveness of the Online Safety Act by holding two public evidence sessions into Ofcom's proposals set out in their 'additional safety measures' consultation. The consultation aims to supplement the codes of practice on illegal harms and protections for children, which came into force earlier this year. Last week (Tuesday 9th Sept) we heard evidence from Beatriz Kira, Bernard Keenan and Dan S. covering various issues ranging from: 🔎 the use of 'proactive technology' such as content identification technology, user profiling technology and behaviour identification technology which raise questions about effectiveness as well as privacy concerns. 🧬 the use of hash matching which involves creating a hash (an identifying series of characters) of a unique piece of content, storing that hash in a database, and using an algorithm to detect attempts to upload the same or similar versions of that content. 📈 proposed changes to recommender systems. Ofcom is suggesting that services should exclude content from recommender systems where there are indications that it is illegal and related to harms of hate, terrorism, suicide or Foreign Interference Offences (FIO) and that content moderation teams must then review such content. 📡livestreaming recommendations including platforms use of human content moderation and user sanctions (such as the NSPCC's suggestion that providers should ensure there are clear consequences for users who consistently break the rules, for example de-ranking of a user’s profile) Dr Kira answered my questions about livestreaming by pointing out that it is "probably one of the most difficult challenges from the perspective of moderation because the potential for harm is in real time." She went on to suggest a potential safety by design approach would be to make livestream content ineligible for algorithmic amplification while it's live. The Online Safety Act's duties are being implemented in stages and Ofcom have pointed out that "online safety is extremely dynamic, with harms, technology and best practice in user safety evolving quickly" necessitating an "iterative approach to the Codes". This approach means stakeholder input and public scrutiny can play an important role. Later today (16 Sept 25), in our second and final session, we will hear evidence from Andy Burrows Rani Govender and my colleague Baroness Beeban Kidron

🚨 𝐇𝐨𝐰 𝐜𝐚𝐧 𝐩𝐥𝐚𝐭𝐟𝐨𝐫𝐦𝐬 𝐞𝐧𝐬𝐮𝐫𝐞 𝐜𝐡𝐢𝐥𝐝 𝐩𝐫𝐢𝐯𝐚𝐜𝐲 𝐬𝐚𝐟𝐞𝐭𝐲 𝐚𝐧𝐝 𝐬𝐞𝐜𝐮𝐫𝐢𝐭𝐲 𝐨𝐧𝐥𝐢𝐧𝐞? 🌐 Article 28 obliges providers of online platforms to put in place appropriate and proportionate measures to ensure a high level of privacy, safety, and security of minors, on their service- but how will this obligation be borne out in practice? 🔔 European Commission has released draft guidelines (open for consultation until 10 June 2025) in furtherance of Article 28(4) of the #DigitalServicesAct to provide measures to ensure the protection of children’s privacy, safety & security online and say: 📢“Children's safety online is our top priority. The draft guidelines we are publishing today under the Digital Services Act will help shape a better and safer online environment for younger users. The wide range of measures will aid the different online platforms in protecting minors' safety, security, privacy and well-being “ 📝 The draft guidelines cover a wide range of measures such as verifying user's age, improving how content is recommended to users to reduce the risk of children being exposed to harmful content, setting children's accounts to private by default, best practices for child-safe content moderation, child-friendly reporting channels and user support, as well as guidance on platforms' internal governance: 🚸Why is Article 28 so crucial (and this consultation document so necessary) Children face real risks online: ⚠️ Illegal/harmful content ⚠️ Grooming & cyberbullying ⚠️ Manipulative algorithms ⚠️ Exploitative nudging ⚠️ Unsafe AI (deepfakes, chatbots) All platforms accessible to minors must take a number of key measures: 1️⃣ Risk Assessment 〰️Platforms must review risks to children using the OECD 5Cs framework 〰️Conduct child rights impact assessments 2️⃣ Age Assurance ❌ Self-declaration not enough ✅ Use age estimation or age verification Provide 2+ methods to avoid exclusion 〰️Use of the upcoming EU 18+ App encouraged 3️⃣ By Design & Default 〰️Set highest privacy/safety defaults 〰️Turn off tracking, autoplay, push notifications 〰️Remove harmful filters, stop persuasive design 4️⃣ Child-Centred Interfaces 〰️Empower children with clear tools to control their experience 〰️Age-appropriate time limits & nudges 〰️Accessible, inclusive interfaces 5️⃣ Safer Recommender Systems 〰️No profiling minors 〰️Avoid risky content loops 〰️Prioritise explicit user signals, not just watch time 〰️Let kids reset their feeds 6️⃣ Ethical Advertising 〰️No ads based on profiling 〰️Clear commercial labelling 〰️Protect against dark patterns & loot box-like features Consultation open until 10 June 2025 see more below ⬇️ and see more here 🔗 https://lnkd.in/eSPawYGB #DSA #ChildOnlineSafety #AgeAppropriateDesign #ChildRights

79% of European teenagers use social media daily. Yet only a fraction of platforms verify their age or prioritise their safety. The European Commission has just unveiled two major steps to protect minors online under the Digital Services Act: 1. New Guidelines on the Protection of Minors These provide risk-based recommendations for platforms to: • Limit addictive features like 'streaks' and 'read receipts' • Empower minors to mute, block, and prevent unsolicited group adds • Prohibit downloads or screenshots of minors’ content • Give children control over what content gets recommended • Default minors’ accounts to private to avoid unwanted contact 2. A Privacy-Preserving Age Verification App (Prototype) Built on the technical specs of the upcoming EU Digital Identity Wallet, this app will: • Let users prove they’re over 18 without sharing their identity or exact age • Be tested in five frontrunner Member States: Denmark, Greece, Spain, France, and Italy • Enable integration into national systems or function independently This initiative is the result of a broad consultation process, including young people, and reinforces the EU’s Better Internet for Kids Strategy. If you work in tech, policy, or platform governance, now is the time to familiarise yourself with these requirements. The online safety of children is no longer optional. It’s enforceable.

MCI responded to Parliamentary Questions from MPs Edward Chia and Sitoh Yih Pin on how Singapore could protect children from harmful online content. (https://lnkd.in/g2cmaY2f) While the high level of digital penetration brings lots of convenience to our daily life, it also presents increasing risks to Singapore users, especially our children. According to a Ministry of Communications and Information (MCI) study last year, half of the parents surveyed reported that their children encountered harmful online content, particularly sexual content and cyberbullying. On the Government’s end, we have enacted legislation and regulations to enhance the safety of users online. For example, • The Code of Practice for Online Safety was introduced last year to require Social Media Services with significant reach or impact to put in place measures to protect users, especially the young, from harmful content. • The Online Criminal Harms Act (OCHA) was introduced last year to address growing criminal activities online, such as terrorism, violence and scams. Two Codes of Practice were also introduced recently under OCHA to require providers of designated online services to put in place measures to proactively disrupt scams and malicious cyber activities. • We are also developing a new Code of Practice for Online Safety - App Distribution Services to address concerns over harmful and age-inappropriate content on apps. Beyond legislation, MCI works with Social Media Services and the community, such as Media Literacy Council, to empower parents to better protect their children online. An example is the Online Safety Digital Toolkit where parents can find parental controls, privacy and reporting tools, as well as self-help resources at https://lnkd.in/gwQJ56GD Members of the public are also encouraged to take part in workshops and family activities organised under IMDA’s #DigitalforLife movement to learn more about healthy and balanced use of technology. (https://lnkd.in/gmH32jMJ) You may have read in the news recently that Ministry of Health (Singapore) and Ministry of Social and Family Development, Singapore (MSF) are also planning to make their advice on device usage and screen time clearer and more definitive and will finalise and announce these initiatives in the coming months. Ultimately, we all care for our young and must do our part to provide a safe online environment for them. #OnlineSafetyTips #StaySafeOnline

🚨 Big milestone for children’s safety online 🚨 Today Ofcom has published new rules that online platforms will need to follow to better protect children. Services will be required by law to assess the risks their platforms pose to children and take steps to keep them safer. That means: ➡️ More choice and support for children. Sites must allow children to indicate what content they don’t like; to accept or decline group chat invitations; to block and mute accounts; and, to disable comments on their own posts. ➡️ Safer feeds. Sites that operates a recommender system and poses a medium or high risk of harmful content must configure their algorithms to filter out harmful content from children’s feeds. ➡️ Effective age checks. The riskiest sites must use highly effective age assurance to identify which users are children. This means they can protect them from harmful material, while preserving adults’ rights to access legal content. ➡️ Fast action. All sites must have processes in place to review, assess and quickly tackle harmful content when they become aware of it. ➡️ Easier reporting and complaints. Children should find it straightforward to report content or complain, and providers should respond with appropriate action. ➡️ Strong governance. All sites must have a named person accountable for children’s safety, and a senior body should annually review the management of risk to children. This is a big step forward in creating a digital world where children can explore, learn and connect more safely – and where the responsibility sits where it should: with the platforms themselves. 📘 Read more about the new rules here: https://lnkd.in/e59zwRrW #OnlineSafety #ChildSafety #DigitalRegulation #Ofcom #TechPolicy

Digital technology offers children incredible opportunities to learn, connect, and express themselves. But it also exposes them to serious risks and harms, including exploitation and abuse. Governments around the world are responding by introducing laws and regulations to make digital spaces safer. Dedicated measures are required to ensure online platform regulation works for children. Our new UNICEF policy brief, Keeping Children Safe Online: Trends in Online Platform Regulation and Emerging Lessons, highlights key developments and lessons from six jurisdictions: Australia, the EU, India, Kazakhstan, South Africa, and the UK. Here are some of the key takeaways: ✅ Businesses must respect children’s rights in the digital environment. Digital products, services, and business practices can negatively impact children’s rights. Governments need a ‘smart mix’ of measures (including laws, policies, incentives, and voluntary standards) to ensure all businesses respect human rights, including children’s rights, online. ✅ Eight common business obligations are emerging across jurisdictions. These include Safety‑ and/or privacy‑by‑design; Impact and/or risk assessments; Age assurance mechanisms; Proactive detection of illegal and/or harmful content; Notice‑and‑action mechanisms; Mandatory reporting of illegal content; Complaints procedures; and Transparency reporting. ✅ Regulation must be grounded in international human rights and children’s rights law. Effective regulation should aim to prevent business‑related abuses of children’s rights, in line with the UN Convention on the Rights of the Child. These rights are universal and indivisible. ✅ Evidence and children’s voices matter. Online platform regulation should be informed by research and meaningful engagement with children and young people. Where evidence is limited, governments may apply the precautionary principle, while avoiding unintended negative rights consequences. The way forward? Online platform regulation is complex and fast‑moving but waiting for a ‘perfect’ model is not an option. Policymakers must act now to protect children and their rights, drawing on lessons from other jurisdictions to design frameworks that are rights‑based and effectively enforced. 👉 Read the full policy brief here: https://lnkd.in/e_5dvMDV #ChildRights #DigitalSafety #PlatformRegulation #TechPolicy #OnlineSafety #Bizhumanrights