Canada Just Banned Under-16s from Social Media and Left the Algorithms Running

Key Takeaways

• Canada introduced Bill C-34, the Safe Social Media Act, on June 10, 2026, banning children under 16 from having social media accounts with platforms that cannot demonstrate adequate child safeguards.
• The bill also imposes new duties on AI chatbot services, including a requirement to immediately redirect users expressing suicidal or self-harm intent toward crisis intervention services staffed by a human.
• Platforms can apply to the new Digital Safety Commission for an exemption from the age ban if they prove their safeguards are sufficient, making this as much a compliance exercise as a hard prohibition.
• The engagement algorithms and addictive design features that researchers link to the mental health harms driving this legislation are not addressed by the ban.

The Bill That Finally Names a Number

Minister of Canadian Identity Marc Miller introduced Bill C-34, the Safe Social Media Act, on June 10, 2026, placing Canada alongside Australia and a growing list of countries that have decided 16 is the line. The bill bans children under that age from social media accounts and brings platforms and AI chatbot services under a new federal framework built around three duties: protect children, act responsibly, and make certain content inaccessible.

The bill text goes into notable detail on the chatbot side. Services meeting the definition must immediately interrupt any conversation in which a user expresses suicidal ideation or an intention to self-harm and redirect them to a crisis service with a live human available. They are also prohibited from posing as a human being in a way likely to deceive users and from using manipulative engagement techniques that encourage emotional attachments in ways that cause social withdrawal.

As for the enforcement, administrative violations carry penalties up to the greater of $10 million or 3% of gross global revenue, with criminal conviction pushing that to 5% or $20 million. A new Digital Safety Commission of Canada handles compliance and audits the digital safety plans platforms must submit and publish.

Platforms can apply to the Commission for relief from the under-16 ban if they can demonstrate adequate safeguards are already in place. Dr. Bolu Ogunyemi of the Canadian Medical Association was sharp about the broader dynamic, saying it was "unacceptable for foreign-owned platforms to continue to get rich at the expense of our children's mental health, privacy, and personal safety." The exemption clause, well, suggests the government is still leaving them a door.

The Part Nobody in Ottawa Wants to Talk About

A ban on account creation does not equal a ban on exposure to harmful algorithmic content. The recommendation systems, the infinite scroll, the autoplay, the engagement optimization engineered to keep users in the app as long as possible – none of it is addressed by Bill C-34. It will still run on every user who remains on the platform, including the adults and any under-16 who creates an account with a parent's birth year, which is not a difficult thing to do.

The compliance burden falls hardest on smaller platforms, exactly as Bluesky's COO warned regarding the UK's social media ban, with a 40-person team facing the same Digital Safety Commission reporting requirements as Meta, which employs thousands of trust and safety staff whose job is building compliance infrastructure around a product architecture that goes unchanged. And while the possibility to get relief from the ban does change things a bit, it’s unlikely to make that much of a difference in any meaningful way.

Meanwhile, Canada is simultaneously running Bill C-22 through Parliament, a surveillance expansion that would mandate backdoor access to encrypted communications and retain metadata on every Canadian for up to a year. A government framing itself as a digital safety champion while building a parallel mass surveillance infrastructure has a coherence problem it has not tried to explain.

And, well, the age verification the bill requires is left to the platforms to figure out, subject to Commission approval, and no jurisdiction running a similar ban has cleanly solved the problem of making verification effective without creating new privacy exposures.

Good Law, Wrong Tool

The AI chatbot duties deserve a genuine concession. Requiring a platform to stop a conversation and direct a user to a human crisis line the moment suicidal intent appears is a practical, specific, and meaningful protection.

And yet, every major study cited in the government's own announcement points to platform design as the driver of harm, specifically engagement-maximizing features, algorithmic amplification of distressing content, and design that exploits psychological vulnerabilities in adolescent development – none of which the bill's framework requires platforms to change.

The Digital Safety Commission will have a lot of compliance reports to read, and platforms are very good at producing those. Whether the Commission will have the appetite or the mandate to challenge the product decisions that sit underneath the reports is a question Bill C-34 does not answer, and it should have.

So, to sum it up, if you want an actual layer of protection while the government figures out what exactly it is that it’s doing, a virtual private network at least lets you control what jurisdictions your traffic appears to come from, and you can get Mysterium VPN with 78% off right now. It won’t fix the algorithm, nor will it make Big Tech pay for the massive harm it’s causing, but at least you will have some actual privacy and security online without being handcuffed by laws that only pretend to be useful.