Hi Name,  

This November in Safety Space, we’ll cover the growing attention given to child safety from regulators and the industry alike, as well as our insights on Trust & Safety standards and the importance of training for consistency in policy enforcement and withstanding regulatory scrutiny. 

Here’s a preview of what you will find in this edition:

🌐 Regulatory Brief: Following in Australia’s footsteps, Europe, Denmark, and Malaysia weigh teen social media bans. The EU court confirmed Amazon’s VLOP status, and Ireland is investigating X’s appeals process. Most VLOPs have now published their latest risk assessments and audit reports — all tracked in our updated DSA database, along with new Trusted Flaggers and ongoing investigations. In the UK, Ofcom issued a £50,000 fine, opened new cases on nudify sites, updated an ongoing suicide-forum case, and released new guidance on protecting women and girls online. 

💻 Inside the Tech: Roblox and Snapchat introduced new age-verification tools. Discord launched a Family Center. OpenAI published a Teen Safety Blueprint on how AI companies should engage with minors. Bluesky expanded its reporting options, while TikTok introduced new in-app resources on violent extremism and launched a Time and Well-being space to support healthier digital habits.

🔵 Insights: We explored in a recent blog post how structured training builds consistent enforcement across teams and is essential in this era of regulatory scrutiny. And, if you missed it, we also published a recap of our latest webinar on operationalising international T&S standards with speakers from Microsoft, DTSP, George Washington University, and Ofcom.

But before we get to the flurry of news, we are excited to share that Julie de Bailliencourt joins us as Tremau’s new Director of Trust & Safety, bringing her vast experience to deepen and expand our T&S expertise in the Advisory & Policy team. With her 25-year track record in building safety infrastructures at tech giants like Meta and TikTok, she’ll bring her deep policy and operational knowledge to support the growth of Nima, our AI-driven T&S orchestration platform, and assist  our clients facing a fast-changing online safety landscape.
Welcome to the team, Julie 👋

🌐🧒 Global Focus: Protecting minors online

• 🇦🇺 From December 10th, the social media ban for under-16s will be fully enforced in Australia, and platforms will need to ensure teens cannot set up accounts on their platforms and that existing accounts are deactivated or removed. 🇪🇺 In Europe, Members of the European Parliament want to implement a similar approach, and they passed a non-binding resolution on age restrictions. Similarly, 🇩🇰 Denmark’s government aims to ban access to social media for anyone under 15. 🇲🇾 Malaysia is also planning to ban social media for under-16s starting in 2026.
• 🌏 The EU, UK, and Australia have formed a trilateral cooperation group focused on age assurance. This group will explore how to further build the technical evidence base with respect to age assurance, and where relevant, how regulators can support independent research in this field.
• 🇬🇧 Itai Tech Ltd was fined in the UK with £50,000 for failing to introduce highly effective age assurance and opened investigations into five platforms operating adult websites for similar breaches. 
• 🇬🇧 Ofcom’s ongoing investigation into an online suicide forum revealed that the site’s IP block for UK users was ineffective and/or not consistently maintained, and the content was still accessible via a "mirror site".

🇪🇺🧑‍⚖️ DSA Compliance & Enforcement

• 🇪🇺 Amazon will keep its VLOP status under the DSA. The General Court of the European Union rejected Amazon’s challenge, stating that the designation and obligations are proportionate and justified for the objective of preventing systemic risks in the Union. 
• 🇮🇪 The DSC, Comisiún na Meán, the Irish Digital Service Coordinator (DSC), is investigating X regarding concerns about its user appeal system for content moderation decisions under the DSA.
• 🇪🇺 Update of our DSA Database: Most VLOPs have released their latest risk assessments and audit reports. You can follow ll of these, along with updates such as new Trust Flaggers and ongoing investigations, in our DSA database.
• 🇪🇺 The European Commission released its first report on the landscape of prominent and recurrent risks on VLOPSEs, as well as reported mitigations. Key findings of recurring risks include the spread of illegal content, threats to fundamental rights, risks to mental health and the protection of minors online. The impact of emerging technologies, such as generative AI, is also highlighted.

🇬🇧📑 OSA's new roadmap and guidance for platforms

• 🇬🇧 Ofcom's updated implementation timeline shows that the register of categorized services & transparency reports is delayed to summer 2026: Under Ofcom's new timeline, platforms need to wait until summer 2026 to know if they will formally be "categorised services" that have additional obligations, notably around transparency. Thresholds for some categories can be as low as 3 million UK monthly active users – many services are expected to be on the list.
• 🇬🇧 New guidance on protecting women and girls online published: This guidance proposes best practices that platforms can adopt to create safer environments, such as risk assessments focusing on harms to women and girls, and conducting usability evaluations on products.
• 🇬🇧 File-storage and file-sharing services receive more detailed guidance on how to comply with the OSA. High-risk services are recommended to use perceptual hash matching technology to proactively detect and remove CSAM. Starting Spring 2026, all services will need to report all detected, unreported CSEA content to the National Crime Agency (NCA).

🔞 Platforms introduce age checks, parental controls, and new AI safeguards

• 🎮 Roblox users will now need to undergo age checks before accessing chat features. The system places users in age-similar groups and limits interactions between adults and minors.
• 🤳 Snapchat will implement age checks in Australia through a software tool that links to the users’ bank accounts. Users will also be able to confirm their age with a selfie or by uploading government-issued ID.
• 🖥️ Discord launched a Family Center, giving guardians more control and visibility over teen activity, including purchases, call time, and top contacts and servers.
• 🤖 OpenAI introduced its Teen Safety Blueprint, which sets suggestions for how AI companies should engage with minors. It calls for age checks and systems that prevent self-harm content, violent material, or help in obtaining illegal substances.

🛡️ Platform Safety and Integrity Updates

• 📲 Bluesky expanded its post-reporting system to have 39 options, allowing users to flag issues more precisely and in response to growing legal obligations. They are also upgrading internal tools so that enforcement decisions are tracked in one place, and users receive updates automatically.
• 🔧 TikTok introduced new in-app resources in Germany for users searching for terms related to violent extremism, created in partnership with the Violence Prevention Network.
• 🧘 TikTok also launched a new Time and Well-being space with tools for relaxation and digital balance, replacing the previous screen time page.

👩‍🏫 Achieving clarity on what “platform safety” means across product, policy, legal, and operations teams remains a challenge. In our latest article, we explore how structured training programmes can turn policy into daily practice by diving into what strong training looks like, who should receive it, and how organisations can scale it.

🖥️ Two weeks ago, we explored global T&S standards and how teams can begin applying them in our webinar “Operationalizing Standards for Trust & Safety”, co-hosted with the Digital Trust and Safety Partnership (DTSP).

David Sullivan (DTSP), Anna Lenhart (George Washington University), Silvia Fukuoka (Ofcom), Pamela Almaguer (Microsoft), and Agne Kaarlep (Tremau) discussed how standards can reduce uncertainty and how teams can map current processes - risk assessments, transparency workflows, governance reviews - to emerging frameworks to identify gaps.

👪 What do guardians and minors want to see in social media safety features? The Center for Democracy & Technology interviewed 45 parents and teens on their perception of age verification, screen-time features, algorithmic feed controls, and parental access. Participants stressed the need for flexibility, transparency, and respect for family dynamics.

🧒 Children's engagement with content controls: Ofcom-commissioned research found that children aged 13-17 often accept default settings rather than adjusting them. This suggests default settings play a major role, and safer options are chosen more often when prompted during sign-up.

🇬🇧 88% of adults agree that the UK government should require tech companies to use an ‘upload prevention’ method in end-to-end encrypted environments to detect images and videos of child sexual abuse: Internet Watch Foundation. Upload prevention works by creating “digital fingerprints” of files, which are compared to a secure database of material that has already been confirmed as child sexual abuse material; if there is a match, then the image is blocked from being uploaded. 

🎮 Prosocial Behaviour in Gaming: The latest episode of Safety is Sexy features Weszt Hart, Head of Player Dynamics at Riot Games, discussing the importance of creating game environments that encourage pro-social behavior and foster positive player interactions.