Ottawa introduces bill to ban social media for kids under 16

Canada’s federal government outlined the details of its newly proposed Safe Social Media Act today, which includes a ban on social media for users younger than 16.

Ottawa announced Bill C-34, which has not yet been debated before parliament, at a press conference on Wednesday afternoon with Minister of Canadian Identity and Culture Marc Miller. The proposed bill includes a swath of potential restrictions on both social media services and artificial intelligence (AI) chatbots, as well as the establishment of a new regulator for online safety.

Under the bill, Ottawa will establish a new regulatory body, dubbed the Canadian Digital Safety Commission (CDSC), that will establish, oversee, and enforce safety standards for both social media services and AI chatbots. That body will be independent and led by members appointed by the Governor in Council, meaning by the Governor General acting on advice of cabinet. The CDSC will be able to issue orders, conduct inspections, and levy fines to companies of up to a maximum of three percent of global revenue or $10 million, whichever is higher, according to government officials.

Most Canadians support ban, poll shows

A March poll from Angus Reid found that three-quarters of Canadians support a youth social media ban, and premiers from provinces including Saskatchewan, Manitoba, and New Brunswick have previously spoken in favour of this type of legislation. Leading Canadian researchers have found that social media has negative impacts on youth mental health, can lead to addictive behaviours, and can lead to other issues such as extortion and cyberbullying.

During a technical briefing on the issue today, government officials said the CDSC could reasonably be up and running within 18 months of Bill C-34 receiving royal assent, with the bill’s regulations—including age-gating—coming online between six to eight months after royal assent.

The proposed legislation outlines several key areas of focus for regulating both AI chatbots and traditional social media sites, defined in the bill as a duty to act responsibly, a duty to make certain content inaccessible, and a duty to be transparent.

On the social media side, a ban would be in place for users under 16 years of age, but social media companies that meet sufficient safety standards—which will be defined at a later date by the CDSC—could see those restrictions lifted. The act will also not apply to private messaging features within social media apps. Chatbots will not be subject to age-gated restrictions under this version of the bill, with officials citing a less robust body of evidence of harm related to chatbots and the ability to make chatbots safer by design.

“Chatbots are different from classical social media. It is very much an evolving playing field, admittedly. They play a function and role that can be very damaging towards kids, but can also play an important function in the educational system and in the AI strategy that we are putting forward,” Miller said. “We’re going to have to keep a close eye on it, but it did not feel appropriate to impose a minimum age restriction at this time.”

Concerns over harms related to chatbot interactions have swirled since the January 2026 mass shooting in Tumbler Ridge, BC, where a teen discussed carrying out violent scenarios with ChatGPT prior to killing nine people. OpenAI, the chatbot’s parent company, was criticized for not doing enough to inform law enforcement about the potential threat.

Both services will see increased scrutiny over online harms, defined by the bill as content related to cyberbullying, self-harm, sexual exploitation, hateful material, content that incites violence, terror or extremism-related content, and non-consensual intimate image sharing. Social media services will be required to more swiftly to remove illegal and harmful material like child sexual abuse material or nonconsensual sharing of intimate images or videos. Under the bill’s framework, companies would have 24 hours to remove that content after it is reported

For AI chatbots like ChatGPT, companies must implement measure to mitigate harmful content generation, and integrate crisis intervention around potential instances of self-harm, suicide, or violence. Companies must also make it more difficult to jailbreak or coerce chatbots into engaging in harmful behaviour or counsel and provide guidelines, and reporting tools for users.

Both types of service providers will also be required to continually assess, mitigate, and report on risks to users, provide tools to flag content, and label bot-driven, harmful, or synthetic deepfake materials. Companies will also need to provide a digital safety plan to the CDSC and release that plan to the public. The bill would also require companies to keep records showing compliance, and to share data with researchers.

Officials said Bill C-34 would apply to most traditional social media and chatbots, but not all. The legislation would target platforms with a large enough user base in Canada, similar to thresholds outlined in Canada’s Online Streaming Act and Online News Act, according to government spokespersons. There would also be the opportunity to regulate smaller platforms if they are deemed to be a vector for harmful behaviour. Large, US firms like Facebook, X, and Snapchat would be included, according to Miller.

“Teenagers can find workarounds”

The Safe Social Media Act is a spiritual successor to Canada’s Online Harms Act, a bill that had been working its way through government for several years before dying while parliament was prorogued in 2025. That legislation received pushback from critics who claimed the act did too much to restrict lawful speech.

This latest bill also comes amid similar bans in countries like Australia, which in December banned kids under 16 from using some social media, and the UK’s 2023 Online Safety Act, which age-gated some online content. The UK is also considering a social media ban for anyone younger than 16.

Advocate groups and some parents have supported a proposed ban, citing concerns over how social media use impacts mental health and safety, as well as its addictive qualities. Critics of this type of legislation, both in Canada and elsewhere, have said such restrictions are largely ineffective.

Jay Goldberg, of the Consumer Choice Centre, criticized the proposed legislation, arguing that verification tools could carry privacy concerns, and pointing to issues with Australia’s ban, where reports have shown that many children are evading the restrictions, either by migrating to different social media platforms not under the ban’s umbrella or by fooling age verification processes.

“Teenagers can find workarounds to access social media, but overall bans can be extremely harmful for those who choose not to, as it disconnects them from support groups, group chats, and mechanisms for staying in touch with distant friends or loved ones,” Goldberg said in an emailed statement.

Josh Tabish, senior director for Canada with tech lobby group the Chamber of Progress, told BetaKit in an email that while he sees parts of the act as an “important step toward making the internet safer,” he doesn’t feel that bans work. “Protecting young people should mean making digital spaces safer, not locking kids out.”

Ottawa has said its version of online safety legislation won’t seek to put perfection before good.

“The efficacy of a law depends obviously on compliance, but we’ve got plenty of laws out there unfortunately that people don’t comply with,” Miller said. “That is why it was so important to have a commission able to interact with industry and to give some teeth to this law so that we didn’t just hope and pray this would be enforced.”

Others critics have highlighted concerns around privacy and data collection related to age-gating that can require users to provide photos of themselves or government-issued ID to access these platforms. Under Bill C-34, the CDSC will be responsible for creating a framework of adequate age-verification protocols for companies to implement. During today’s technical briefing, government officials said criteria for devising those methods would include that it is efficient, collects a minimum amount of data, destroys that data when it’s no longer needed, and complies with the commission’s framework.

Feature image courtesy of Marc Miller via X