Comments on the OPC Consultation on Artificial Intelligence

On January 28, 2020, the Office of the Privacy Commissioner of Canada (OPC) published its Consultation on the OPC’s Proposals for ensuring appropriate regulation of artificial intelligence (Consultation Paper). The Consultation Paper sets out several proposals for how the federal Personal Information Protection and Electronic Documents Act (PIPEDA) could be reformed, in the words of the OPC, “in order to bolster privacy protection and achieve responsible innovation in a digital era that involves artificial intelligence (AI) systems.” The document also invites privacy experts to validate the OPC’s understanding of how privacy principles should apply and whether its proposals would be consistent with the responsible development and deployment of artificial intelligence, calling for responses to be submitted by March 13, 2020.

The Consultation Paper considers the perspectives of other bodies that have treated the issues raised by the various proposals at length, including the OECD, the IEEE and the UK Information Commissioner’s Office among others. This makes the document substantial, and commenting on the Consultation Paper in its entirety is not feasible in a short post.

There is no question that the recent convergence of cheap, on-demand computing resources, very large data collections and the development of machine learning platforms makes it timely to consider legal reforms that better address the promise and the risks surrounding the latest wave of developments in AI.

That said, any consideration of this topic must begin with a caveat: the term “artificial intelligence” has a long history in computer science, cognitive science and philosophy, and its meaning has become rather elastic. This can be useful for marketing but hinders legal analysis.

This content has been updated on May 2, 2024 at 13 h 46 min.