-
DESCRIPTIONArtificial intelligence is transforming the world faster than any technology in decades, but without intentional design, it risks deepening the access divide for blind and low-vision users. AIMAC (AI & Machine Learning Accessibility Community) has launched an evaluation framework creating shared standards, best practices, and accountability tools that ensure AI systems work for everyone. This session explores how AIMAC enables developers to evaluate how effectively large language models generate accessible code through an open-source benchmark, with the Accessibility Leader Board at aimac.ai publicly displaying comparative scores of different LLMs based on how accessible their generated HTML outputs are.
Speakers
-
Moderator: Karae Lisle, Chief Executive Officer, Vista Center for the Blind and Visually Impaired
-
Joe Devon, Entrepreneur and Accessibility Advocate, GAAD / A11y Audits
-
Emon McErlean, VP & Global Head of Digital Accessibility & Globalization, ServiceNow
-
-
SESSION TRANSCRIPT
