The Intervention Journey A Roadmap to Effective Digital Safety Measures 2025

Example policy-related interventions TABLE 3 Intervention Description Example organization(s) External User appeal process A formal process enabling users to challenge and request a review of content moderation decisions, such as removals or account suspensions.YouTube Allows users to appeal video takedowns and account suspensions, reviewed by human moderators. Policy enforcement mechanismsA framework that imposes penalties for violating platform policies, such as issuing warnings, strikes, temporary suspensions or permanent bans.Discord Informs users when they have broken rules, what actions have been taken and how it may impact their overall account standing Equity and inclusion strategiesPolicies aimed at ensuring platform policies and moderation practices protect vulnerable and marginalized groups from disproportionate harm while promoting inclusivity.TikTok Implements policies to protect minority groups from hate speech and discriminatory content, promoting an inclusive online community. External engagement with experts and civil societyCollaborating with external organizations, such as researchers, NGOs and civil society groups, to improve policies and content moderation practices and understand risks related to online harms.Meta Oversight Board An independent panel of experts that reviews content moderation decisions and advises on policy improvements. Internal Risk intelligence capabilitiesInternal systems and capabilities designed to detect, assess and mitigate emerging risks, such as cyber threats, misinformation and other harmful content, guiding platform adjustments.Google Uses sophisticated risk detection systems to monitor threats like misinformation, harmful content and child safety risks, allowing for proactive policy changes. Assurance and audit processesReviews and audits of content moderation practices to ensure compliance with platform policies, legal standards and ethical guidelines, maintaining consistency in enforcement.Meta The independent Data Transparency Advisory Group provides an assessment of Facebook’s content moderation. Risk triage and escalation processesProcesses used to assess, categorize and escalate cases of online harms (e.g. terrorism, CSAM) based on severity, ensuring the most serious issues are handled by appropriate teams, including legal and trust and safety teams.YouTube Uses risk escalation for sensitive content like terrorism or CSAM, where specialized teams make final decisions. Employee support systems, including for trust and safety teams and othersPlatforms that provide mental health support, counselling and stress management resources for employees, especially those exposed to harmful content, like trust and safety and content moderation teams. Concentrix Monitor psychological safety of moderators and provide ongoing training and practical resources. Source: Google Transparency Report. (n.d.). Appeals; Discord. (2024). Discord Warning System; TikTok Safety Center. (2025). Inclusion and Belonging Guide; Oversight Board. (n.d.). Improving how Meta treats people and communities around the world; Google Safety Centre. (n.d.). Content safety; Meta. (2019). An Independent Report on How We Measure Content Moderation; Google Transparency Report. (n.d.). Featured policies; Concentrix. (n.d.). Safe Spaces: How to Create Psychological Safety in Content Moderation. The Intervention Journey: A Roadmap to Effective Digital Safety Measures 32