Bill C-34 Safe Social Media Act

Read Full Bill Text Here

C-34 An Act to enact the Digital Safety Act and the Digital Safety Commission of Canada Act and to make consequential amendments to other Acts

Short Title: Safe Social Media Act

Bill Type: House Government Bill

Bill Sponsor: Minister of Canadian Identity and Culture and Minister responsible for Official Languages

Status: Introduced (1st Reading) — June 10, 2026. This Bill hasn't passed yet.

How would YOU vote? Scroll down to vote and comment below.

WHO GAINS POWER

  • A new federal regulator — the Digital Safety Commission of Canada — is created with broad authority to investigate, inspect, fine and order social media platforms, chatbot services and other online services to comply with this Act
  • The Commission can issue guidelines, conduct hearings, designate inspectors, enter places (with warrant), compel testimony and access electronic data
  • The Minister can require the Commission to hand over any information it holds, direct it to conduct reviews and request reports on any matter within its mandate
  • The Governor in Council (Cabinet) sets which platforms are regulated, what the minimum user thresholds are, which AI systems are exempt and when the Act comes into force — all by regulation, without a parliamentary vote
  • Cabinet can also designate entirely new categories of online services as regulated if it decides they pose a "significant risk of harm to children"

⚠️ "Significant number of users" is not defined in the Act — Cabinet sets the threshold by regulation and can change it at any time without returning to Parliament.

⚠️ Which platforms are covered is determined entirely by regulation — Cabinet can add or remove platforms from the regulated list without a vote.

⚠️ The entire Act comes into force by Order in Council — Cabinet decides when (and whether) any part of this law takes effect.

WHO LOSES POWER

  • Social media platforms (Facebook, Instagram, TikTok, X, YouTube, Reddit, adult content sites, live streaming services) must implement harm-mitigation measures, age verification, content labeling, user blocking tools, flagging tools and resource persons — or face fines
  • Chatbot services (AI companions, conversational AI) must interrupt interactions and redirect users to crisis services if a user expresses suicidal ideation or intent to harm; must not impersonate humans or licensed professionals
  • Users who post content that sexually victimizes a child or constitutes non-consensual intimate content can have that content made inaccessible within 24 hours — before any finding of wrongdoing
  • Platforms must preserve removed violent or terrorism content for one year and hand it over to law enforcement on request
  • Operators lose the ability to keep their internal safety data, moderation volumes and research findings private — all must be disclosed in publicly available Digital Safety Plans

WHO GAINS MONEY

  • The Commission is funded in part through cost-recovery charges levied directly on regulated platforms — Cabinet sets the charge amounts by regulation
  • Researchers and educators accredited by the Commission gain access to platform data inventories for research projects
  • Legal and compliance professionals will see significant demand as every regulated platform must build out internal safety infrastructure

WHO LOSES MONEY

  • Platform operators face administrative monetary penalties of up to the greater of $10 million or 3% of gross global revenue per violation — and each day of a continued violation is a separate violation
  • Criminal conviction (indictment) can result in fines up to the greater of $20 million or 5% of gross global revenue
  • Platforms must fund age-verification systems, content moderation infrastructure, resource persons, digital safety plans and ongoing compliance reporting
  • Cost-recovery charges are a debt to the Crown — recoverable in Federal Court

THE CATCH

  • No proactive search required — platforms are not required to scan their own content for harmful material (s. 12), but regulations can require technological measures to prevent child sexual abuse material from being uploaded (s. 126(i))
  • "Harmful content" is broadly defined — it includes not just child sexual abuse material and non-consensual intimate images, but also content that "foments hatred," "incites violence," bullies a child or induces self-harm — all subject to platform-level moderation duties
  • Content can be made inaccessible within 24 hours of being flagged, before any independent review — the operator decides, subject to a representations process it also controls
  • The Commission is not a competent witness in any proceeding other than a prosecution under this Act — meaning its internal deliberations are shielded from outside scrutiny
  • Chatbot exemptions are set by regulation — Cabinet can carve out entire categories of AI systems from coverage without returning to Parliament
  • The Act coordinates with Bill C-16 (Protecting Victims Act) — if that Bill also passes, mandatory reporting obligations are folded in automatically

⚠️ "Content that foments hatred" is defined by reference to the Canadian Human Rights Act — what qualifies is determined by the Commission's guidelines, which it writes itself, subject only to a requirement to consider freedom of expression.

⚠️ Digital Safety Plans must be publicly posted — platforms must disclose moderation volumes, flagging data, research findings and resource allocation, but can withhold trade secrets and confidential commercial information at their own discretion.

[Source: Bill C-34, 1st Session, 45th Parliament — ourcommons.ca]