Artificial intelligence and the law in North America: a transatlantic analysis

This article represents the second part of our comparative transatlantic analysis of the legislative initiatives taken to regulate artificial intelligence (AI) on both sides of the Atlantic Ocean, i.e. in the European Union (EU) on the one hand, and in Canada and the United States, on the other. To learn more about what is being done at this level in the EU, we refer you to the first part of this text: Artificial intelligence and law in Europe: comparative transatlantic analysis “.

Artificial Intelligence and the Law in the United States

In the United States, on December 10, 2021, the US Federal Trade Commission (FTC) published a Statement of Regulatory Priorities, demonstrating the intention of this government agency to initiate the procedure to propose the development and adoption of laws and regulations on issues such as privacy, security and “algorithmic decision-making”. ), which are not currently subject to any federal law. In particular, the FTC is considering using its regulatory power to limit practices that are not too restrictive in terms of confidentiality and invasions of privacy, as well as to ensure that decision-making processes based on algorithms (i.e. i.e. using AI) do not result in discrimination. There are still few details on the next steps of this FTC initiative, but it is expected that significant changes will take place in 2022 in the field of AI in the United States.

It is also relevant to look at what the various American states have adopted in terms of legislation regarding AI. General laws or resolutions were proposed in 17 states in 2021, and were passed in four states:

  • Alabama: Law Al SB 78 (Technology, Alabama Council on Advanced Technology, estab., to advise Governor and Legislature, members, duties) aims to establish the “Alabama Council on Advanced Technology and Artificial Intelligence”. The latter’s role will be to review the work of the governor and legislator of Alabama, as well as that of all other interested parties, and to advise them on the use and development of advanced technologies and AI in the ‘State.
  • Colorado: CO SB 169 (Act Concerning Protecting Consumers From Unfair Discrimination in Insurance Practices) prohibits insurers from using any external data concerning their customers, as well as algorithms or predictive models using external data sources, which could give rise to discrimination based on race, skin color, national or ethnic origin, etc.
  • Illinois: Amends the Artificial Intelligence Video Interview Act (IL HB 53) amends the Artificial Intelligence Video Interview Act and provides that employers who fully use AI to determine whether a person should be invited for an in-person interview are required to collect certain demographic data and report it to the Department of Commerce and Economic Opportunity. The latter has an obligation to analyze the data transmitted to it in order to determine whether the use of AI in this sector gives rise to racial discrimination. He must then report it to the Governor and the General Assembly of Illinois.
  • Mississippi: MS HB 633 (Mississippi Computer Science and Cyber ​​Education Equality Act) requires the State of Mississippi to set up training courses relating in particular to AI and “machine learning” in schools at all levels of education. ‘State.

AI and the law in Canada

Canada has yet to pass AI-specific legislation, either federally or provincially. Nevertheless, at the end of 2020, the Canadian government proposed a new law: the Consumer Privacy Protection Act (CPPA). This Canadian law on personal data, which is part of a desire for transparency and accountability of economic players, will replace the current Personal Information Protection and Electronic Documents Act (PIPEDA). Among other things, the transparency requirement will be strengthened, and organizations will have to provide all necessary information about how they use automated decision-making systems (i.e., using algorithms that can be assimilated to AI) to make predictions or decisions about users. Users will have the right to request further explanation of these procedures.

The PIPEDA amendment also opened the door to revising provincial privacy laws in Ontario, British Columbia, Alberta and Quebec. It is certain that AI will be a subject that will be taken into consideration when developing new regulations, as demonstrated among others by the Ontario White Paper “ Modernizing Privacy in Ontario ”, published as part of the public consultation in connection with the modernization of privacy protection in Ontario. This document emphasizes, among other things, the importance of regulating profiling and automated decision-making practices.

In terms of amending current provincial privacy laws, Quebec has led the way by adopting the Bill n°64 : An Act to modernize legislative provisions relating to the protection of personal information (hereinafter “Bill 64”) on September 21, 2021. With the aim of modernizing the framework applicable to the protection of personal information, Bill 64 amends a number of laws, in particular the Data Protection Act. personal information in the private sector (hereinafter “LPRPSP”). Like the CPPA, it imposes an obligation on public bodies and companies that use personal data in order to make a decision based exclusively on automated processing thereof to inform the person concerned. In addition, Bill 64 provides a framework for biometric procedures. For example, an organization that wants to implement a biometric database and system will need to do a privacy impact assessment prior to any project involving biometric data.

In conclusion, the legal landscape relating to AI, both nationally and internationally, will undergo many changes in the coming years. It is essential for companies to keep up to date on this subject, since this means that new obligations will also be imposed on them.

A law firm practicing English-speaking common law in Great Britain, Canada and the United States as well as French-speaking civil law in Quebec and France, S. Grynwajc also has particular expertise as’GDPR lawyer, in the field of personal data protection. Consequently, if you wish to ensure that your organization complies with the new obligations incumbent on it, do not hesitate to contact us, we will be delighted to accompany you!

We would love to thank the author of this write-up for this awesome material

Artificial intelligence and the law in North America: a transatlantic analysis

Explore our social media profiles as well as other pages related to them