This is part one of a three-part series.
With software such as ChatGPT, artificial intelligence (AI) has become an increasingly hot topic, not only in tech but across all industries – including legal – on a global scale.
On May 4, 2023, BLG Partner François Joli-Coeur was joined by Marc Etienne Ouimette, Global Lead, AI Policy (AWS) at Amazon and Michael Bahar, Partner at Eversheds Sutherland to discuss different approaches for regulating AI in Canada, the U.S. and Europe, preparing compliance programs, and the latest in AI litigation and enforcement.
The below is a summary of the upcoming AI legal framework. To get full details on strengths and weaknesses of legislative approaches, opportunities for interoperability and what businesses can do to prepare, check out the full 30-minute webinar recording or skim the transcript*.
AI legal framework across jurisdictions
As part of Bill C-27, the Canadian government introduced new legislation, the Artificial Intelligence and Data Act (AIDA), to set the foundation for the responsible design, development and deployment of AI systems. AIDA’s focus is on high impact systems, which are outlined in the AIDA companion document.
In the U.S., government regulators are ensuring they are mitigating risk from AI tools before capitalizing on the opportunities. Existing laws and regulations are going to be applied vigorously by regulators like the Federal Trade Commission, Consumer Financial Protection Board, the U.S. Department of Justice, and others.
With the EU’s Artificial Intelligence Act, the EU seems to want to establish the gold standard in which all other nations orient their own legislation. Two significant points about the law, which is set to come into force by the end 2023:
- It will have extraterritorial reach and like the General Data Protection Regulation (GDPR), it’s reach could radiate outwards to other places, such as Canada and the U.S.
- The fines are sizable with the idea that companies internalize costs.
Approaches to legislation differ based on jurisdiction. The two approaches taken are horizontal approach and vertical approach:
Aims to cover all AI solutions and classify them as either high risk, no risk or prohibited, and establish a set of obligations associated with high-risk deployments.
Along with a few other countries, Canada is trying to take this approach with Bill C-27.
Rather than a top-down approach regulating all of AI, this approach aims to go industry by industry or department by department.
This is the approach being taken by the U.S. and U.K.
To better assist companies on their digital transformation journey, BLG’s 3-part webinar series covers artificial intelligence, the metaverse and the Internet of Things. If you have any questions about the legal framework around AI, or the topics covered in this series, please reach out to any of the authors or key contacts listed below.
*Recording and transcript are available in English only.