Inclusive design lies at the heart of accessible and effective AI
By Lara Davis, Director of Communications, Marketing and Digital at Business Disability Forum.
Artificial intelligence (AI) is quickly becoming smarter as it moves towards being applied strategically, but who is this technology really being designed by and for? A recent poll of 1,032 disabled adults we conducted with research firm Opinium suggests that disabled people should be involved in conception and development from the very beginning.
The poll found that 40 per cent of disabled adults believe AI would be more accessible if they were directly involved in the design, development and testing of AI products. Respondents were asked for their views on how to improve AI accessibility for disabled people and inclusive design came out on top – even before technical fixes or post‑launch adjustments.
Other top answers given included creating more user-friendly interfaces (38 per cent), greater availability of information about how AI can support disabled people (37 per cent) and more support to help disabled people get started with AI (36 per cent).
Inclusive design is not about special features added at the end of development, but is rather about embedding accessibility into products, services and systems from the start to ensure they reflect the real needs, preferences and lived experiences of disabled people.
Why inclusive design matters in AI
AI systems increasingly shape how people communicate, work, learn, shop and access everyday services. If disabled people are not included early in development, AI risks replicating the very barriers it promises to remove through biased data, incompatible interfaces, or decisions made without input from those with lived experience.
BDF’s poll responses reflect this concern but also highlight an opportunity. Alongside inclusive design, disabled people’s views were sought on how AI could help them now and in the future. Findings show that over a third believe that AI tools can help disabled people by improving communications (38 per cent) and improving online experiences (34 per cent).
Other benefits that were mentioned include improved access to healthcare information and communication (33 per cent), better access to education, improved accessibility of digital content (each 32 per cent), support for independent living (31 per cent), improved customer experience (25 per cent) and better access to employment and in the workplace (24 per cent).
Inclusive design bridges these gaps. By involving disabled people throughout the AI lifecycle, developers can anticipate problems before they become barriers, avoid harmful assumptions and build systems that work across a wider range of abilities and contexts.
Better outcomes for all
AI must not be style over substance and must instead rely on inclusive choices being made upfront. These include features such as voice interfaces that work with speech differences, AI tools that integrate with assistive technology, or content systems that present information in multiple formats.
However, the research also found that 1 in 5 disabled adults (20 per cent) do not believe AI can help disabled people now or in the future. This scepticism reflects a lack of trust and accessibility in current systems and reinforces why inclusive design is essential.
Inclusion is a responsibility, not an afterthought
There is the potential for AI products and tools to make a radical and positive difference to disabled people’s lives, but there is also the risk that disabled people could be left behind. With AI developing at pace and 1 in 4 people experiencing disability at some point in their lives, this is not an issue that we can afford to overlook.
Businesses, AI developers and employers must actively consult disabled people to ensure they are involved in the design, development and testing of AI products to assess their suitability. Disabled people also need better access to information and advice about AI, more generally.
Lucy Ruck, who leads our Tech Taskforce, reinforces the belief that inclusive design is all about smart design.
“AI has the capacity to transform lives, but only if we get inclusion right from the start. Making sure that disabled people are active participants in shaping this technology isn’t just the right thing to do, it’s how we build AI that genuinely serves everyone.”
What this means in practice
For businesses and AI developers, it means fully involving and consulting disabled people from the outset through to the AI lifecycle. This includes clearly communicating accessibility features, ensuring compatibility with assistive technologies and applying ethical judgement and human oversight, so AI doesn’t introduce new barriers or reinforce bias.
For employers, inclusive design must be embedded in AI strategy. Accessibility and disability inclusion should be core considerations that are non-negotiable in AI use. Disabled employees should be consulted throughout procurement, testing and rollout, and AI should support, rather than replace, human decision-making. Accessible training is essential so that all staff understand both how to use AI tools and why accessibility matters.
As AI continues to reshape our world, designing with disabled people, not for them, is how the technology will become more accessible, more effective and more human.
More practical guidance is available in our AI resources.
Read the press release on the new poll.