Nearly one in four people in the European Union has a disability. For them, a looming legal deadline—the European Accessibility Act (EAA), which comes into force in June 2025—is set to reshape the digital world. For business leaders, it represents more than a compliance race; it’s a fundamental test of their artificial intelligence (AI) strategy.
The EAA mandates that products and services, from banking apps to e-commerce sites, must be accessible. While this follows in the footsteps of legislation like the Americans with Disabilities Act (ADA), its arrival in the AI era raises a more profound question: How can you build accessible services on top of AI models that are often inherently inaccessible?
The AI Accuracy Gap: A Data-Driven Problem
Disability is often underrepresented or incorrectly categorised in the public datasets used to train AI models. This creates an “accuracy gap,” where systems fail to reflect the lived experience of people with disabilities, leading to inherent bias and inaccessibility. AI image generators that don’t understand mobility aids or voice assistants that can’t parse atypical speech patterns are not just technical flaws; they are the direct result of a data-driven blind spot.
The EAA’s deadline is forcing a reckoning with this problem. But a handful of tech leaders are not just aiming for compliance; they are proactively using AI and data to solve it.
The Pioneers: How Tech Leaders Are Building for Accessibility
“Accessibility Is Part of Our DNA,” Says Tim Cook, CEO of Apple
Apple has announced new accessibility features launching later this year. “Accessibility Nutrition Labels” will add a new section to App Store product pages that highlights accessibility features within apps. This includes support for VoiceOver, Larger Text, and captions.
Microsoft Is Making AI More Inclusive
Since 2024, Microsoft has been working with Be My Eyes, a mobile app that connects users with vision impairments to sighted assistants, to make AI models more inclusive for the over 340 million people worldwide who are blind or have low vision. With unique video datasets to promote effective training, the company aims to gain a deeper understanding of the challenges the visually impaired community faces. This effort could bridge the “accuracy gap” limiting the usefulness of AI models.
Adobe Is Helping the Industry Meet Standards
Content management systems like Adobe Experience Manager offer built-in tools for accessibility compliance, so organisations can more easily meet necessary standards. These platforms help users to create and maintain websites that meet Web Content Accessibility Guidelines (WCAG) standards and automate the testing of digital content for accessibility issues.
Partnerships with Tech and Advocacy Drive Action
At the Google Cloud Summit in London earlier this year, Jeff Dodds, CEO of Formula E, announced the introduction of audio race reports powered by Google Cloud’s generative AI (GenAI) to produce vivid, multilingual descriptive audio summaries for each E-Prix race. The initiative aims to offer a dynamic experience for blind and visually impaired spectators. Furthermore, in collaboration with the Royal National Institute of Blind People (RNIB), Formula E and Google Cloud will conduct user testing sessions—with a comprehensive rollout of more accessible features scheduled for Season 12.
From Compliance to Competitive Advantage
This isn’t just a compliance issue; it’s an architectural one. The true test of a modern AI system is not how it performs in the average case, but how it serves the edge cases. The companies highlighted above are not merely retrofitting accessibility; they are using it as a catalyst for innovation.
They understand that building for accessibility leads to better products for everyone. An interface that works for a user with low vision is often a cleaner interface for all users. A voice assistant that understands diverse speech patterns is a more robust assistant overall. In the AI era, accessibility is no longer just a feature; it is a measure of quality, resilience, and strategic foresight.
A Leader’s Playbook to Building More Accessible Digital Journeys
- Audit Your Stack – Use tools like Axe or Lighthouse, and include manual testing with real users.
- Train Teams Across Functions – Train developers, designers, marketers, and customer service teams on accessibility best practices.
- Design Inclusively from Day One – It’s harder to retrofit accessibility. Bake it into design from the start.
- Make Content Perceivable – Add alt text to images, ensure keyboard navigation, provide transcripts for videos, and use readable fonts and contrast.
- Publish Accessibility Statements – Communicate how your products meet standards and actively ask for user feedback.
