CAN-ASC-6.7: Accessible and Equitable Physical Artificial Intelligence Systems
Information
Technical committee members
- Nancy McLaughlin, Senior Policy Advisor on Accessibility, Canadian Radio-television and Telecommunications Commission
- John Willis, Senior Program Advisor, OPS Accessibility Office, Centre of Excellence for Human Rights
- Toufic Sbeiti, Director of Social Media and Visual Communications, Immigration, Refugees and Citizenship Canada
- Jutta Treviranus (Chairperson), Director of the Inclusive Design Research Centre and Professor, OCAD University
- Gary Birch, Executive Director, Neil Squire Society
- Lisa Liskovoi, Senior Inclusive Designer and Digital Accessibility Specialist, Inclusive Design Research Centre, OCAD University
- Clayton Lewis, Professor, University of Colorado
- Tara Connolly, Assistant Director Research and Development, Accessibility Institute, Carleton University Julia Stoyanovich, Associate Professor and Director, Tandon School of Engineering, New York University
- Anne Jackson, Professor, Seneca College
- Kave Noori, Artificial Intelligence Policy Officer, European Disability Forum
- Mia Ahlgren, Human Rights and Disability Policy Officer, Swedish Disability Rights Federation
- Sambhavi Chandrashekar (Vice-Chairperson), Global Accessibility Lead, D2L Corporation
- Julianna Rowsell, Senior Product Manager, Product Equity, Adobe
- Hala Anwar, AI Research Leadership, Adobe Labs
- Kate Kalcevich, Head of Accessibility Innovation, Fable
- Saeid Molladavoudi, Senior Data Science Advisor, Statistics Canada
- Merve Hickok, Founder, President and Research Director, Alethicist.org, Center for AI and Digital Policy, University of Michigan
The purpose of CAN-ASC-6.7 is to develop a standard that goes above mandatory minimum technical specifications and produces equity-based technical requirements.
There are common areas where people with disabilities may face barriers related to artificial intelligence technologies, including physical artificial intelligence systems. These include, but are not limited to:
- Misinterpretations by physical artificial intelligence systems that do not take into account the diverse bodies, movements, communication methods, behaviours, or support needs of people with disabilities;
- The inability to anticipate, avoid, interrupt, stop or override the AI-enabled physical actions or responses;
- Biases or exclusions in operation that produce unreliable, or exclusionary outcomes for people with disabilities.
To address these barriers, this standard will help ensure that physical artificial intelligence systems do not amplify harms, biases, or disadvantages and support the equitable participation of people with disabilities in the use, operation, and benefits of physical AI. It will then establish technical requirements to support physical AI, including but not limited to:
- Methods for how physical AI systems can account for diverse bodies, movements, communication methods, behaviours, or support needs of people with disabilities;
- How users can understand, anticipate, avoid, interrupt, stop, override, or get assistance with AI-enabled physical actions or responses; and
- How to address biases in operation to reduce unreliable, exclusionary, and disadvantageous outcomes.
This standard will align with all relevant Accessibility Standards Canada standards.