Canadian civil society organisations, alongside AI and human rights experts, have urged in an open letter, the Canadian government to uphold the integrity of the first binding international treaty on AI. This treaty, known as the Council of Europe Framework Convention on AI, Human Rights, Democracy, and Rule of Law, is currently undergoing negotiations, with concerns raised about the potential inclusion of a private sector carve-out that could weaken rights and limit the accountability of technology companies.

The call is clear: Canada should commit unequivocally to safeguarding human rights from AI risks and harms, irrespective of the source. The delegation's alternative stance would not only violate the Canadian Charter of Rights and Freedoms but also breach the fundamental constitutional principle that the same rules should apply to all, including the private sector.

As a contributing member to several Council of Europe treaties, including the Framework Convention on AI, Canada's recent position in the negotiations is causing concern, said the signatures of the letter. Reports suggest that the Canadian delegation is pushing to limit the Convention's scope by removing provisions that establish safeguards for private-sector AI systems, a stance vehemently opposed by civil society organisations and experts.

The potential impact of a watered-down Convention raises alarms, as it could provide insufficient protection to individuals facing powerful AI systems prone to bias, manipulation, and the destabilisation of democratic institutions. This stance also questions Canada's commitment to protect human rights, democracy, and the rule of law domestically, especially in the context of ongoing negotiations of Bill C-27, addressing consumer privacy protection and AI.

Canada, known for its leadership in global AI initiatives, is urged to reconsider its position. The current stance could undermine efforts to safeguard fundamental rights, democratic values, and the rule of law. It contradicts Canada's involvement in international initiatives such as the Global Partnership on AI, where it has championed human-centred values, fairness, transparency, and accountability.

The potential fallout from Canada's current position is extensive. It could damage the transatlantic relationship, jeopardise data transfer agreements with the European Union, and impact the equivalence decision regarding Canada's protection of personal information. To mitigate these risks, the Government of Canada is urged to reassume a leadership role on trustworthy AI policy and instruct its delegation to the Council of Europe to stand against any private sector carve-out.