Artificial Intelligence was an undeniably hot topic in 2023.
Following the launch of ChatGPT in November 2022, AI quickly became an active part of the collective consciousness.
It has also raised new, or perhaps renewed, controversy about the use and implications of AI, with many expressing concerns over a variety of issues, including ownership and copyright infringement, security threats and privacy issues, the elimination of jobs and other ethical considerations.
Should AI Be regulated?
With so much potentially at stake, the question becomes what, if anything, should be done.
One solution may be government regulation, a proposition many seem to support.
“The use of AI creates new and more pronounced risks of harm to individuals and groups. An AI-specific statute can help prevent these harms from materializing,” explained Nic Wall*, a senior associate at Torys LLP in Toronto, whose practice focuses on privacy, cybersecurity and data governance. "While the common law has shown some ability to adapt to new technology, it often only does so slowly and retrospectively. There is also a benefit to the industry in having concrete legislative parameters instead of hoping their adherence to best practices satisfy various regulators or withstand scrutiny in litigation.”
Osgoode Hall Law School professor Pina D’Agostino, who specializes in intellectual property, technology and innovation law and policy, also believes that AI should be regulated. “As a legal scholar and someone who cares about society, I think we need to have rules in place that are clear, transparent, ethical and understandable to the public and to the people who need to be governed by those laws,” she affirmed. “We need to know what the rules of engagement are, whether you're a company, a consumer or a member of the public."
Mark Doble, CEO of Alexi**, a Toronto-based legal tech company that uses proprietary generative AI to execute legal research and produce memos for licensed legal professionals in Canada and the US, is open to regulation but with a caveat. “AI has the potential to have an immensely positive impact on society and humanity at large. It also has the potential to do harm,” he observed. “I do think that there is a role for some regulation that can mitigate the potential harms, but it is critical that any regulation does not hinder continued research and development so that we may realize the full benefits. This will be difficult to get right.”
Canada Makes First Attempt To Regulate AI With Legislation
There is currently no legislation in force specific to AI in Canada, despite the fact that it is home to 20 public AI research labs, 75 AI incubators and accelerators, 60 groups of AI investors from across the country and over 850 AI-related start-up businesses.
However, on June 15, 2023, the Canadian government made its first foray into AI regulation with Bill C-27, the Digital Charter Implementation Act, 2022. The Bill introduces the country’s first piece of legislation intended to regulate the development and use of AI systems, called the Artificial Intelligence and Data Act (AIDA).
The government describes the legislation as “a framework intended to ensure the proactive identification and mitigation of risks in order to prevent harms and discriminatory outcomes while recognizing the unique nature of the AI ecosystem and ensuring that research and responsible innovation are supported.”
The Ministry for Innovation, Science and Economic Development summarizes the approach under AIDA as follows:
- Building on existing Canadian consumer protection and human rights law, AIDA would ensure that high-impact AI systems meet existing expectations with respect to safety and human rights. Regulations defining which systems would be considered high-impact, as well as specific requirements, would be developed in consultation with a broad range of stakeholders to ensure that they are effective at protecting the interests of the Canadian public while avoiding imposing an undue burden on the Canadian AI ecosystem.
- The Minister would be empowered to administer and enforce the Act to ensure that policy and enforcement move together as the technology evolves. An office headed by a new AI and Data Commissioner would be created as a centre of expertise in support of both regulatory development and administration of the Act. The role would undergo gradual evolution of the functions of the commissioner from solely education and assistance to also include compliance and enforcement once the Act has come into force and the ecosystem has adjusted.
- Prohibit reckless and malicious uses of AI that cause serious harm to Canadians and their interests through the creation of new criminal law provisions.
The government has indicated that its consultation process will result in the provisions of AIDA coming into force no sooner than 2025.
In September 2023, the Federal government further announced a “Voluntary Code of Conduct on the Responsible Development and Management of Advanced Generative AI Systems.” The Code is intended to temporarily provide Canadian companies with common standards and to enable them to demonstrate, voluntarily, that they are developing and using generative AI systems responsibly until formal regulation is in effect.
Several leading AI companies have announced that they will sign the voluntary Code, although others have indicated they will not support it.
Concerns Remain
While committee work has begun on the Bill, its language and content have not been finalized. In fact, in early October 2023, the Minister for Innovation, Science and Economic Development, François-Philippe Champagne, sent a letter to the Chair of the committee indicating support for certain amendments to further clarify important aspects of the law in response to input from various stakeholders.
Others share similar concerns with the law as presently proposed.
“AIDA seems well-intentioned and is a good starting point, but needs some work,” Wall said. “Stakeholders from all sides have been pushing for amendments and clarifications. Industry is requesting certainty and clarity on the scope of application, the meaning of key definitions and who is responsible for complying with the AIDA’s substantive requirements.”
Wall also notes that human rights and civil society groups are indicating that the Bill doesn’t go far enough in regulating AI or mitigating impacts on communities. Additionally, he observes that there’s a tension between a need for certainty and a need to avoid creating a statute that is overly prescriptive and inflexible, which risks stifling innovation. “Achieving the right balance of certainty and flexibility will no doubt be challenging.”
Doble has similar apprehensions. “I have significant concerns that AIDA will negatively impact Canadians' ability to innovate and deliver real value with AI,” he said. “It is too easy to focus on the theorized negative outcomes, as seen in movies and the other common media AI tropes. The positive outcomes unfortunately don't make for such great stories, but they are far more likely if we let people be creative and innovative. This imbalance in how we collectively view AI has a significant risk, it seems to me, to lead to over-regulation and the inadvertent impediment to progress.”
Ultimately, whether the legislation actually achieves its stated goals, and whether it does so in a way that avoids stifling AI innovation or businesses more generally, will depend a lot on its execution, Wall concluded.
* Recognized in the 2024 edition of Best Lawyers: Ones to Watch™ in Canada in Privacy and Data Security Law and Technology Law.
** Disclaimer: the author is currently employed at Alexi as a legal researcher.
Sara Collin is a Québec-based lawyer, specializing in legal writing, editing, research and translation throughout Canada.
