Vector Institute, Alberta Machine Intelligence Institute and Mila – Quebec Artificial Intelligence Institute

Les informations de ce site Web à été fournie par des sources externes. Le gouvernement du Canada n'assume aucune responsabilité concernant la précision, l'actualité ou la fiabilité des informations fournies par les sources externes. Les utilisateurs qui désirent employer cette information devraient consulter directement la source des informations. Le contenu fournit par les sources externes n'est pas assujetti aux exigences sur les langues officielles et la protection des renseignements personnels.

September 2021

Consultation on a Modern Copyright Framework for Artificial Intelligence and the Internet of Things

Joint Submission of Vector Institute, Alberta Machine Intelligence Institute and Mila – Quebec Artificial Intelligence Institute

Canada is poised to be a leader in the development and commercialization of artificial intelligence (AI). We have world renowned researchers and academic institutions who are at the leading edge of deep learning, machine learning and other forms of AI. Our excellence and leadership in the knowledge, creation, and use of AI has positioned Canada to achieve transformative economic growth, while improving the lives of Canadians.

Achieving this success will require a policy framework that supports AI activities. Fundamentally, this means policies that promote access to data and remove barriers to AI development and commercialization.

AI development requires sifting through, analyzing, and understanding a wide variety of materials which enable computers to aid humans. Teaching a computer to safely drive a vehicle or diagnose an illness requires large data sets. This   may mean   analyzing   tens of   millions of photographs, articles, and other works to train software to interpret text, recognize patterns, and make predictions.

Copyright may be impacted by AI research because analyzing data to teach a computer often requires making an incidental technical copy of a lawfully accessed work. That incidental copy shouldn’t require additional permissions to enable its use for analytics. However, the ambiguity around copyright law is often enough to dissuade researchers from making lawful analytical uses of machine readable information.

Many countries with the ambition of becoming world leaders in AI have or are taking steps to eliminate uncertainty about the copyright implications of AI development. If Canada is to keep pace with these and other countries, it is necessary that the Copyright Act be amended to expressly allow for the reproduction of lawfully accessed works to facilitate information analysis (also referred to as text and data mining). It is not an overstatement to predict that a failure to do so will have a chilling effect on AI development and commercialization in Canada – both because access to data required to perform research will be curtailed and because the best AI talent and investment capital will migrate to jurisdictions with more favourable laws.

Amending the Copyright Act will not undermine the interests of content owners. Once lawful access to a work is obtained, it should not matter for copyright purposes whether the work is read or viewed by an individual or by a computer. In both cases, the work is accessed to acquire information or insight that is not protected by copyright. Additionally, enabling information analysis will not impact on a content owner’s capacity to control – and charge for – access to its work.

The signatories to this submission represent Canada’s leading AI-research and academic institutions and AI companies. As with all Canadians, we have a direct stake in successfully leveraging AI to achieve transformative economic growth, while improving lives across the country. Because of this, we strongly recommend that Canada’s copyright law be amended to include an exception to infringement for information analysis. Implementing this recommendation will enhance our shared vision of making Canada a leader in the development and commercialization of AI.

Garth Gibson, President & CEO
Vector Institute

Cam Linke, CEO
Alberta Machine Intelligence Institute

Valérie Pisano, CEO
Mila – Quebec Artificial Intelligence Institute