Canada introduced Bill C-34, the Safe Social Media Act, on June 10, 2026. It bans anyone under 16 from having a social media account and puts new legal obligations on both social platforms and AI chatbot services. This is not a soft guideline: platforms face fines of up to 3% of their global revenue or 10 million Canadian dollars per violation, whichever is larger.
The bill draws three clear duties for platforms. They must protect children by building safer products by design. They must make certain content permanently inaccessible, specifically material that sexually exploits children and non-consensual intimate content including deepfakes. And they must act responsibly by labelling AI-generated content, providing blocking and reporting tools, and submitting publicly available safety plans to a new regulator called the Digital Safety Commission of Canada.
AI chatbots are not age-blocked under this bill, but they are regulated. Services like ChatGPT will be required to have clear procedures for when a user talks about harming themselves or others, and to be transparent about when and how they escalate those situations. This requirement was directly shaped by the Tumbler Ridge school shooting in February 2026, where eight people died, including six children. OpenAI had flagged and banned the shooter's account eight months before the attack after internal systems detected discussions of violence, but never alerted police. CEO Sam Altman publicly apologized in April 2026. Families of victims subsequently filed lawsuits in California.
This is the Canadian government's fourth attempt at this kind of legislation since 2021. A previous version died when Parliament was prorogued in early 2025. This time, the bill avoids the contentious criminal law changes that sank its predecessor, narrowing the focus to platform accountability and child safety to make it easier to pass.
The bill still needs to clear the House of Commons and Senate. Given typical parliamentary timelines and the additional regulations that follow after it becomes law, the rules are unlikely to affect users in practice before late 2027 or early 2028.
The Australian experience is a useful reality check. Australia passed a near-identical law in December 2025. Within weeks, platforms had removed or restricted over 4.7 million accounts judged to belong to under-16s. Yet by early 2026, researchers found that roughly 75% of 14 to 15-year-olds in Australia were still not complying, largely because platforms had simply failed to identify and remove their accounts in the first place. Australia's regulator launched formal investigations into Facebook, Instagram, Snapchat, TikTok, and YouTube in March 2026 for suspected breaches. The law is being tested in Australian courts too, with Reddit and a civil liberties group both challenging it.
For any business that operates consumer-facing platforms or tools used by teenagers, the direction of travel is now clear across multiple major markets. Canada, Australia, the UK, France, and others are all moving toward holding platforms legally responsible for what happens to children on their services, rather than leaving the burden on parents. The fines are real, the regulators are being built, and the political support is broad: over two-thirds of Canadians support the ban, according to recent polling.
What this means practically is that platforms serving Canadian users will need to build and document age verification systems, content removal processes, and crisis escalation procedures. The exemption pathway exists, but it requires proving to the regulator that safeguards are genuinely in place, not just promised. Companies that operate in multiple countries would be wise to treat Canada's rules as the new baseline, not an edge case.