"Accessibility is not a bolt-on feature. It is a fundamental part of how we build products." - Satya Nadella, Microsoft
Current gap
India must address a legal gap regarding technology. Current laws define accessibility as the absence of barriers, but software requires a proactive approach. We need a tiered definition to empower citizens through affirmative action rather than just avoiding exclusion.
Instant stage of entry
The first level requires compatibility with assistive tools from day one. Foundational models must support diverse vocal patterns and regional languages. This ensures interfaces provide entry rights, allowing everyone to participate in the digital economy without technical hurdles.
Active facilitation
Accessibility should move toward active facilitation where AI simplifies access compared to traditional means. Features like vision assistants reduce user load when navigating portals. This stage enhances interaction quality through tools that proactively assist users with various impairments.
Systemic inclusion
Accessibility translates into systemic inclusion by eliminating historical biases. AI must restructure the environment to ensure participation as equals in society. Inclusion should be baked into training sets to ensure models are universally accommodative and promote social equity.
Proactive and adaptive standards
India should adopt frameworks recognizing proactive duties and individualized adjustments. This creates a self correcting loop where models are inclusive by design and adaptive by nature. By respecting dignity, our code can realize the barrier free environment promised by law.
Summary
This article proposes a tiered definition of AI accessibility to move beyond negative legal standards. By focusing on entry, facilitation, and systemic inclusion, India can ensure future technology is proactive and adaptive, ultimately respecting the dignity of every citizen through inclusive design.
More insights on inclusion here
Food for thought
Can AI achieve systemic inclusion if the data used to train it lacks representation from the disability community?
AI concept to learn: Ex ante accessibility
This refers to the practice of building inclusive features into an AI model from the very moment of its creation. Instead of fixing (repairing) barriers after users face issues, this proactive approach ensures that the technology is universally usable right from its initial deployment.
[The Billion Hopes Research Team shares the latest AI updates for learning and awareness. Various sources are used. All copyrights acknowledged. This is not a professional, financial, personal or medical advice. Please consult domain experts before making decisions. Feedback welcome!]
