A question that has quietly gathered force over the past decade is now pressing itself into the centre of public debate in Canada and beyond: when the products built by the world’s largest technology companies cause measurable harm to people, who bears responsibility? The platforms that mediate so much of modern life — shaping how Canadians consume news, how children experience adolescence, how workers find employment, and how communities organize themselves — have long operated in a regulatory environment that has struggled to keep pace with their reach and influence.
The analogy to other industries is instructive, even if imperfect. Pharmaceutical companies face liability when their drugs cause unforeseen injury. Automakers are held to safety standards enforced by the state. Food producers must meet labelling requirements designed to protect consumers who cannot independently verify what they are ingesting. The underlying logic in each case is the same: when a product enters the lives of millions of people, the entity that designed and distributed it carries a duty of care that the market alone cannot enforce. The question now being asked — with increasing urgency by regulators, civil society organizations, and ordinary citizens — is why the technology sector has so far been treated differently.
This is not a purely abstract debate. It has concrete stakes for Canadians, whose federal government has been working, haltingly and amid considerable political turbulence, to develop a legislative framework for artificial intelligence through the proposed Artificial Intelligence and Data Act. That effort reflects a genuine recognition that the absence of clear rules creates real costs — costs borne disproportionately by the most vulnerable users of these systems.
At the same time, the Canadian investment landscape is signalling where private capital believes the future lies. Recent announcements from RBC, Wittington Ventures, and Power Sustainable point toward a deepening commitment to AI-related ventures, suggesting that the financial sector is moving quickly to position itself within an ecosystem that remains, for now, only loosely governed. The scale of these investments is not incidental to the accountability question — it reflects the degree to which AI infrastructure is becoming embedded in institutions that touch everyday Canadian life, from banking to retail to energy.
The federal government’s parallel effort to boost AI literacy among Canadians deserves more attention than it typically receives. There is something quietly important about the state investing in the capacity of its citizens to understand the technologies that govern increasing portions of their lives. Informed citizens are better positioned to demand accountability, to participate meaningfully in regulatory consultations, and to recognize when a system is producing outcomes that are unjust or opaque. Literacy, in this sense, is not merely educational — it is civic infrastructure.
Critics of stronger tech regulation often warn that heavy-handed rules will stifle innovation, drive investment elsewhere, or place Canadian companies at a disadvantage relative to American and Chinese competitors operating in less restrictive environments. These are not frivolous concerns, and a serious regulatory framework must grapple with them honestly. But the argument that accountability necessarily conflicts with innovation rests on a premise worth scrutinizing: that the harms currently externalized onto users, communities, and democratic institutions are simply the unavoidable price of technological progress. That premise is contestable, and increasingly contested.
Canada’s parliamentary tradition and its federal structure offer both resources and complications for addressing this challenge. Jurisdiction over some dimensions of technology regulation sits with Ottawa, while others — consumer protection, property and civil rights, aspects of privacy — involve the provinces in ways that require coordination rather than unilateral action. Quebec, which has moved further and faster than most jurisdictions in North America on privacy legislation through Law 25, offers a useful case study in what assertive provincial action looks like, and what its limits are when the platforms being regulated operate globally.
Indigenous communities, whose data, languages, and cultural materials have frequently been incorporated into AI training sets without meaningful consent or benefit-sharing, represent a dimension of the accountability question that Canadian public debate has been slow to centre. The principle of data sovereignty — that communities have the right to govern how information about them is collected, used, and shared — is increasingly articulated by Indigenous organizations, and it sits in productive tension with the assumptions built into most commercial AI development. Any serious Canadian framework for tech accountability will need to reckon with this, not as an afterthought, but as a foundational matter.
None of this resolves easily. The question of how to hold technology companies accountable for harm is genuinely difficult, involving contested evidence about causation, complex trade-offs between competing values, and institutions — courts, legislatures, regulatory agencies — that were not designed with these problems in mind. But difficulty is not the same as impossibility, and the fact that the question is hard does not justify deferring it indefinitely while the harms accumulate.
What is clear is that the current default — in which platforms bear minimal legal liability for the consequences of their design choices, their algorithmic amplification decisions, and their data practices — is itself a policy choice, not a natural state of affairs. It was arrived at through specific legislative decisions, lobbying campaigns, and judicial interpretations, and it can be revisited. The more important question is not whether tech giants should be held responsible for the harm caused by their products, but how that responsibility should be defined, allocated, and enforced in a way that is proportionate, evidence-based, and genuinely protective of the public interest. Canada has both the institutions and the democratic tradition to take that question seriously. Whether it will summon the political will to do so remains to be seen.
