Contributing to AI Standards 

A Canadian Artificial Intelligence (AI) standard will help guide the ethical use of AI systems and ensure people with disabilities aren’t negatively impacted by them. This standard will promote equitable access to AI and the participation of people with disabilities in artificial intelligence systems development. The lack of collaboration with the disability community has historically led to the creation of barriers in society, and it’s critical to avoid this from happening in AI systems. 

We are at an inflection point where AI can either accelerate inclusion or exclusion. AI has the potential to enhance assistive technologies, opening up new avenues for independence and career opportunities. It can also prevent certain people from accessing opportunities and filter them out of the hiring process. The responsibility lies with all of us to ensure innovation is beneficial and not harmful. This standard aims to ensure that AI innovation will benefit all members of society. 

Expert insight on accessibility and disability 

Kate has worked in the field of digital accessibility since 2001 and identifies as hearing disabled. Her career has spanned the technical work of making websites accessible, inclusive policy development, teaching accessible design and development, thought leadership in accessibility accommodations, and creating practices to help large organizations mature their approach to accessibility. She brings a unique combination of lived experience, governance and policy background, as well as business and innovation insights to the committee. 

AI’s potential to help people with disabilities 

Fable’s mission is to empower people with disabilities to participate in, contribute to, and shape society. The potential for AI to improve quality of life for people with disabilities is high, if done right. AI tools can describe objects and images to people who are blind, support people with hearing loss via AI-generated captions and help people who are neurodivergent with tasks like writing and staying organized. AI can also collect and interpret data to help with building more inclusive products and services for people with disabilities. The possibilities are endless.

AI’s potential risks for people with disabilities 

The flip side of this potential is the risk of further excluding people with disabilities from accessing:  

• equitable education, if AI-based performance tools can’t account for student differences
• employment opportunities, if AI recruitment tools are modelled on a company’s previous hires they can perpetuate low disability representation
• affordable health insurance, if AI predicts a higher cost of care  
• and many more critical aspects of society

The risks are varied and need to be considered throughout many parts of AI development: 

Data – lack of diverse data accurately representing a variety of disabilities leads to biased outcomes for AI applications. 

Privacy – either exposing private information about people with disabilities inappropriately or denying them access to necessary accommodations. 

Tools – developer tools used to create AI systems, or the AI systems themselves, not being usable by people with disabilities. 

Collaboration will lead to inclusive AI 

Now, more than ever, it is critical to include the perspectives of people with disabilities in developing artificial intelligence systems.  

According to our survey on AI and disability, only 7% of respondents believe that there is adequate representation of people with disabilities in the development of AI tools. Fable’s community of assistive technology users want to participate in AI development and testing. 

Fable was founded on the principle of including people with disabilities into the product development lifecycle. AI products will become safer, more robust, and more usable when accessibility and inclusion standards are followed. We’re excited to help organizations better understand how to innovate inclusively. 

Learn more about Innovation at Fable.