The defining trust and safety trends for 2026 are global age restrictions on social media, the rise of AI companion apps, a content authenticity crisis driven by synthetic media, and a structural shift from reactive moderation to proactive, systems-level risk management. WebPurify, an IntouchCX company, asked eight senior leaders where the industry is heading.

Why this matters now

Trust and safety, the discipline of protecting users from harm on digital platforms, is moving from isolated policy experiments to global norms. Australia’s ban on users under 16 took effect on December 10 and covers platforms such as Instagram, TikTok, Snapchat, YouTube and X. Malaysia, Denmark and the European Parliament are moving in the same direction, and Ireland has signaled age verification plans of its own.

At the same time, generative AI is changing the volume and speed of harm. Some forecasts cited in the report suggest close to 90% of web content could be AI generated by 2026. Technologies that were once optional are becoming regulatory requirements, and the question of who holds power over moderation decisions is being rewritten by new laws, new tools and new user expectations.

Key insights from the whitepaper

Will under-16s really be banned from social media in 2026?

The report predicts an early-stage global normalization of youth age restrictions. Ailís Daly, Head of Trust & Safety, EMEA at WebPurify, warns that the policy has a blind spot: displaced teens.

“We are removing teens without building what comes next.”

Without investment in offline youth programs and safe digital alternatives, she argues, regulators risk solving one safety problem while creating a loneliness crisis.

Why are AI companion apps the next major safety risk?

Teens removed from social media will turn to tools that talk back to them. The European Parliament’s resolution already groups AI companions with social media in its proposed minimum age. Alexandra Popken, SVP of Trust & Safety and AI Services at WebPurify, notes that chatbots tend to be overly sycophantic and make mistakes, some of them deadly. The report expects regulators, parents and the public to demand a new safety standard for AI that imitates relationships, built through red teaming before launch rather than after an incident.

Will AI replace human content moderators?

According to WebPurify’s experts, no. AI has supported moderation for almost 20 years, and LLMs can now supercharge it with contextual understanding. Cost, infrastructure and organizational maturity still limit production-scale deployment. Moderating AI also creates new human work: labeling training data, QA on model outputs, prompt refinement and red teaming.

How should platforms handle AI slop and synthetic content?

AI slop refers to low-quality, mass-produced AI content. The harder problem is high-quality synthetic media that is genuinely difficult to tell from reality. AI generated does not automatically mean harmful, which leaves trust and safety teams in gray zones. Current labeling relies heavily on self-reporting, and provenance standards such as C2PA (the Coalition for Content Provenance and Authenticity) are being adopted piecemeal across companies and jurisdictions.

Is trust and safety becoming a business function?

Yes. AJ Brown of The Brand Safety Institute expects trust and safety investment to be judged by revenue protection, advertiser confidence, user retention and cost reduction. Sophie Walsh, Director of Trust & Safety at Depop, predicts teams will be measured by harm reduction rather than enforcement volume, and embedded alongside product, engineering, legal and communications.

What’s inside the full whitepaper

• All 13 trend predictions in full, including NCII prevention, user dispute resolution under the EU Digital Services Act, toxic-platform disengagement, anticipatory risk management for ride-sharing and real-world platforms, and the regulatory turning point on violence against women and girls (VAWG)
• Commentary from named experts at WebPurify, Appeals Centre Europe, The Brand Safety Institute, Psychgeist, Glitch, Bolt, Depop and Stop NCII advisory
• The three-part model for AI savviness: safety technology, digital literacy and cultural norm-setting, and why AI can surface the signal but humans still make the call in trust & safety
• How privacy-preserving hashing stops NCII before upload, plus the tools every platform with an image upload button should know about
• The 2026 regulatory convergence: the TAKE IT DOWN Act, the UK VAWG strategy, the UN Cybercrime Treaty and Ofcom’s Additional Safety Measures, including the estimated scale of platforms affected by mandatory hashing
• Bolt’s approach to anticipatory detection across the full ride lifecycle, from account creation to post-trip activity
• A blueprint for systems-level trust and safety: continuous risk sensing, pre-launch risk reviews, safety-by-default safeguards and governance structures
• Guidance on building external expert networks for specialized risks such as extremism

Plan your 2026 trust and safety strategy

The full report gives you the regulatory timeline, the operational playbooks and the expert reasoning behind each prediction, so you can prioritize investment before the landscape shifts again. Download it to brief your leadership team with confidence.