Online Safety Legislation in Canada: Bill C-34
After the Online Harms Act died in January 2025, the federal government introduced Bill C-34 in June 2026 to enact the Digital Safety Act. The online safety legislation would impose duties on regulated social media and chatbot services to act responsibly, protect children and be transparent, with 24-hour takedown duties for the worst content and a new Digital Safety Commission.

Key takeaways
- Bill C-63, the Online Harms Act, died on prorogation in January 2025.
- Bill C-34 was introduced on 10 June 2026 and passed second reading on 29 October 2026.
- Regulated chatbot services that simulate sustained human-like relationships are included.
- Child sexual victimization and non-consensual intimate content must generally be made inaccessible within 24 hours.
- No general obligation to proactively search content is imposed.
- Penalties of up to the greater of $10 million and 3% of gross global revenue are proposed.
Online safety legislation in Canada: where things stand
Canada has debated online safety legislation for several years. The federal government's first comprehensive attempt, the Online Harms Act in Bill C-63, was introduced in February 2024 and died when Parliament was prorogued in January 2025. In the current Parliament, the government introduced Bill C-34 on 10 June 2026. It would enact the Digital Safety Act and the Digital Safety Commission of Canada Act, with the overall short title Safe Social Media Act. The bill passed second reading on 29 October 2026 and, as of November 2026, was being considered in committee in the House of Commons. It is not law, and it may be amended, but it shows the shape that federal online safety legislation is likely to take.

This insight summarizes the main features of the proposed online safety legislation as introduced at first reading, how it would apply to social media and chatbot services, the regulator's powers, and what platforms operating in Canada should do now. It is general information, not legal advice.
Purpose and approach
Like comparable online safety legislation abroad, the bill aims to promote the safety of persons in Canada, particularly children, reduce harms caused by harmful content online, and ensure that operators of regulated services are transparent and accountable. Rather than prescribing how every piece of content must be handled, it imposes duties on operators to act responsibly, protect children and be transparent, supported by regulations, guidelines and oversight by a dedicated regulator. The bill expressly provides that nothing in it requires an operator to proactively search content to identify harmful content, although regulations may require technological means to prevent uploads of content that sexually victimizes a child or revictimizes a survivor.
Seven categories of harmful content
The bill defines harmful content as seven categories: intimate content communicated without consent (including AI-generated sexual depictions of identifiable people), content that sexually victimizes a child or revictimizes a survivor, content that induces a child to harm themselves, content used to bully a child, content that foments hatred, content that incites violence, and terrorism or violent extremism content. Each category has its own definition. For businesses, the definitions matter because they set the scope of risk mitigation duties and of the more stringent takedown duties, which apply only to the first two categories.
Which services are covered
The bill distinguishes three types of regulated services. A social media service is a website or application accessible in Canada whose primary purpose is to facilitate interprovincial or international online communication among users by enabling them to access and share content. A service that does not allow users to communicate content to a potentially unlimited number of users is not a social media service, and private messaging features are excluded from the duties. A social media service is regulated if it meets a user threshold set by regulation for its type, or if it is designated by regulation as a regulated service.
A chatbot service is an AI system publicly accessible in Canada that uses a natural language interface to provide adaptive, human-like conversational responses, is capable of simulating a sustained human-like relationship with a user, such as friendship, an intimate relationship or therapeutic support, and generates responses not fully predetermined by its developer or operator. As with social media, regulated chatbot services are those meeting user thresholds or designated for significant risk of harm. The Governor in Council could also extend the duty to protect children and the duty to be transparent to other online services by regulation.
| Duty | Applies to | Key elements |
|---|---|---|
| Act responsibly | Regulated social media services | Adequate measures to mitigate exposure to harmful content, user tools to block and flag, labelling of synthetic content |
| Act responsibly | Regulated chatbot services | Mitigate risk of harmful content and harmful behaviours, crisis intervention measures |
| Protect children | All regulated services | Design features set by regulation, age measures for pornographic content, possible minimum age for accounts |
| Make certain content inaccessible | Regulated social media services | Child sexual victimization content and non-consensual intimate content, generally within 24 hours |
| Be transparent | All regulated services | Digital safety plans submitted to the Commission, records |
The duty to act responsibly
For regulated social media services, the duty to act responsibly would require operators to implement measures adequate to mitigate the risk that users will be exposed to harmful content, to give users tools to block other users and flag harmful content, and to label synthetic content, meaning audio or visual representations made by electronic means, including AI, that are likely to be mistaken for authentic recordings.
For regulated chatbot services, operators would have to implement measures adequate to mitigate the risk that the chatbot communicates harmful content to a user, measures to ensure intervention in crisis situations, and measures to mitigate the risk that the chatbot engages in certain harmful behaviours.
This is a risk-based standard. What is "adequate" will be informed by regulations, Commission guidelines and the operator's own assessment of risks on its service. Operators should expect the Commission to examine the evidence behind their choices.
The duty to protect children
Children, defined as persons under 18, are the bill's central concern. Operators of all regulated services would have to integrate design features set out in regulations. Where an operator has reasonable grounds to suspect that a regulated service provides access to pornographic content, it must implement age-verification or age-estimation measures adequate to mitigate the risk that children are exposed to it. The Commission would assess adequacy, including whether the measures are effective and whether personal information is collected or used only for age assurance purposes. The bill would also allow regulations requiring minimum-age restrictions for having an account on regulated social media services.
These requirements interact with privacy law. The Office of the Privacy Commissioner's Children's Privacy Code calls for risk-based, privacy-protective age assurance and high-privacy defaults for services likely to be accessed by children. Platforms should design age assurance and child safety features that satisfy both regimes. See our insight on children's privacy and age assurance.
Content that must be made inaccessible
The most stringent duty applies to two categories on regulated social media services: content that sexually victimizes a child or revictimizes a survivor, and intimate content communicated without consent. When the operator identifies such content, or a user flags it, the operator would have to make it inaccessible to persons in Canada and notify the relevant users within 24 hours, or another period set by regulation, subject to procedures allowing representations and review.
Persons in Canada could also complain to the Commission, which could order content made inaccessible if the complainant had used best efforts to seek recourse from the operator first. These provisions sit alongside existing Criminal Code offences relating to child sexual abuse material and non-consensual distribution of intimate images.
Transparency and digital safety plans
Operators would have to submit digital safety plans to the Digital Safety Commission, describing the measures taken to comply with their duties, the effectiveness of those measures and related information, so the Commission can assess compliance. The plans must not contain personal information. Operators would also have record-keeping obligations and duties to preserve certain harmful content. Members of the public could make submissions to the Commission about the measures operators take. Transparency requirements of this kind tend to drive internal investment in measurement and documentation, because the plans will be scrutinized.
The Digital Safety Commission and enforcement
Part 2 of the bill would establish the Digital Safety Commission of Canada, with powers, duties and functions under the Digital Safety Act, including issuing guidelines, receiving digital safety plans, handling complaints, holding hearings, conducting inspections and imposing penalties. The maximum administrative monetary penalty for a violation would be the greater of $10 million and 3% of the person's gross global revenue in the preceding financial year. Certain offences, including obstruction of the Commission or inspectors, could lead on indictment to fines of up to the greater of $20 million and 5% of gross global revenue. These figures reflect the bill at first reading and may change.
Global platforms already subject to the EU Digital Services Act or the UK Online Safety Act will recognize the structure: risk assessments, child safety design, user tools, transparency reporting and a regulator with revenue-based penalties. Build a Canadian layer onto existing compliance programs rather than starting from scratch, but check definitions and timelines carefully because they differ.
Implications for AI chatbot providers
The inclusion of chatbot services is one of the most significant changes from the previous bill. Providers of AI companions, virtual friends and wellness or support chatbots should assess whether their products could simulate a sustained human-like relationship, and therefore fall within the definition if user thresholds are met or a designation is made. Customer service bots are less likely to qualify, but businesses should confirm this as regulations are developed. Crisis intervention, protections for minors and controls on manipulative or harmful behaviours are likely to be central to compliance. See our insight on AI chatbots and liability.
Online safety legislation, free expression and privacy
Any online safety legislation raises Charter questions, particularly around freedom of expression and privacy. The bill's design choices, including the absence of a general proactive monitoring obligation, the exclusion of private messaging, the focus on systemic duties rather than individual content decisions for most categories, and the limitation of takedown duties to two categories, reflect efforts to address those concerns. Committee study may refine these features, and any final law may be challenged. Platforms should follow the debate closely, because amendments could change the scope of their obligations.
Preparing for online safety legislation
Even before enactment, platforms that may be regulated can take sensible steps. Estimate Canadian user numbers for each service type, since thresholds will be set by regulation. Map existing trust and safety measures against the seven categories of harmful content and the proposed duties.
Review child safety design, age assurance and parental tools against both the bill and the OPC's Children's Privacy Code. Assess capacity to make flagged child sexual victimization content and non-consensual intimate content inaccessible within 24 hours, with notice and review processes. Prepare to label synthetic content. Begin building the data needed for a digital safety plan, and identify executives accountable for Canadian online safety compliance.
Related measures and private members' bills
Bill C-34 is not the only online safety legislation proposed in the current Parliament. Other bills address young persons' online access to pornographic material, deepfakes and online fraud, and some provinces are considering their own measures relating to minors online. Private members' and Senate public bills often do not become law, but they signal political priorities and may influence amendments to government legislation. Businesses should track the government bill as the main vehicle while watching for overlapping requirements.
How TRW & Co helps
We advise social media, gaming, AI and other online platforms on Canadian online safety legislation, children's privacy, age assurance, content moderation processes, transparency reporting and engagement with Parliament and regulators. See our communications and media and privacy and data protection pages or contact us.
Frequently asked questions
Is Bill C-34 law?
No. It was introduced on 10 June 2026, passed second reading on 29 October 2026 and was at committee in the House of Commons as of November 2026. It may be amended before enactment.
Which platforms would be covered?
Regulated social media services and regulated chatbot services that meet user thresholds set by regulation or are designated for significant risk of harm, plus other online services if regulations extend certain duties to them.
Would platforms have to monitor all content?
No. The bill states that it does not require proactive searching of content, although regulations could require technology to prevent uploads of child sexual victimization content.
What are the penalties?
Administrative monetary penalties of up to the greater of $10 million and 3% of gross global revenue per violation, and higher fines for certain offences, as proposed at first reading.
How does this relate to children's privacy rules?
Platforms will need to align child safety design and age assurance with both the proposed Act and the OPC's Children's Privacy Code, which calls for risk-based, privacy-protective age assurance.
Sources
General information only, not legal advice. References to statutes, courts and authorities describe published law and practice at the date shown.


