The end of the interface? What AI means for accessibility and inclusion

For decades, the way we interact with digital services has been relatively stable. We open an app, navigate a website, tap buttons, complete forms, and move through structured journeys that organisations have carefully designed for us. Accessibility, rightly, has focused on making those journeys usable for everyone - ensuring that websites and apps can be navigated with a keyboard, understood by screen readers, and experienced by people with a wide range of needs.

But a quiet shift is underway. And it may prove to be one of the most significant changes to digital inclusion we’ve seen.

Increasingly, people are not navigating interfaces at all. They are simply asking for what they want.

“Book me a train to Manchester tomorrow morning.”
“Find me the best deal on home insurance.”
“Order my usual groceries.”

AI is beginning to handle the rest.

This is more than a new feature. It is the early shape of a fundamentally different model of interaction - one where conversation replaces navigation, and where intelligent agents act on our behalf across multiple services. For disabled people, this presents both extraordinary opportunities and very real risks.

From navigating systems to stating intent

Traditional digital accessibility has focused on reducing friction within interfaces. Clear layouts, logical structure, accessible forms, and consistent navigation all play a crucial role. However, conversational AI shifts the burden away from the user needing to understand a system, towards the system understanding the user. That is potentially transformative.

For someone with a visual impairment, the need to interact with complex, often inaccessible elements could be dramatically reduced. For someone with cognitive challenges, interpreting confusing layouts or navigating a multi-step process may no longer be necessary. For people with mobility impairments, the physical effort of interacting with multiple controls could be replaced by a single spoken or typed instruction.

The implications are particularly significant in financial services. Managing money is one of the cornerstones of independent living, yet banking, insurance, savings and investment platforms can still present barriers for many disabled customers. An AI assistant that can securely help someone check their balance, compare products, make payments or switch providers through a simple conversation could reduce friction dramatically. Getting this right could unlock greater financial independence for millions of people. Getting it wrong could create entirely new forms of exclusion from essential services.

In its best form, this model removes layers of interaction that have historically created barriers - but that promise depends entirely on how these systems are designed.

A new accessibility challenge is emerging

If AI becomes the primary interface, accessibility does not disappear - it moves. Instead of asking “is this website accessible?”, we may need to ask:

• Can the AI understand diverse ways of communicating intent?
• Does it work equally well with different speech patterns, accents, and assistive technologies?
• Can users trust it to act accurately and transparently on their behalf?
• Will it display any of the biases or prejudices already experienced by disabled users?
• Does it provide feedback in ways that are perceivable and understandable to everyone?

These are not trivial questions. They represent a shift from accessible interfaces to a truly inclusive interaction model.

For example, conversational systems that rely heavily on natural language processing may struggle with non-standard speech patterns, including those associated with certain disabilities. Similarly, if an AI agent makes decisions on behalf of a user without clearly explaining its reasoning, this could create new forms of exclusion - particularly for people who rely on predictability and clarity.

There is also a risk that, in removing visible interfaces, we remove the cues and structure that many users depend on. A well-designed form can guide someone step-by-step. A purely conversational interaction may not. In other words, while AI has the potential to simplify, it can also obscure.

Inclusion must be designed into AI from the outset

This is why inclusion cannot be an afterthought in the age of AI. If anything, it becomes more critical. Organisations developing AI-driven experiences need to think carefully about:

• How users initiate and control interactions
• How choices and options are presented
• How errors are handled and corrected
• How users can review and confirm actions before they are taken
• How systems adapt to individual preferences and needs over time

Personalisation, when done well, could be a powerful enabler. AI systems could learn how an individual prefers to communicate, what level of detail they need, and how they like information to be presented. For disabled users, this could mean truly tailored experiences that go far beyond what static interfaces can offer today. But without deliberate design, personalisation can just as easily reinforce exclusion.

The role of websites and apps is not going away - yet it is tempting to see this shift and assume that traditional digital channels will become irrelevant. In reality, that is unlikely to happen any time soon.
Websites and apps remain the foundation of digital services. They are where content lives, where transactions are completed, and where legal and regulatory requirements are met. They also provide the structured, predictable environments that many users rely on.

Even as AI layers on top of these systems, the underlying experience still needs to be robust, accessible, and inclusive. In fact, AI systems themselves often depend on well-structured, accessible content to function effectively. So while the interface may evolve, the need for accessible digital foundations does not diminish. If anything, it becomes more important.

A moment for leadership, not hesitation

For organisations committed to inclusion, this is not a moment to step back. It is a moment to lead. That means continuing to invest in accessible websites and apps, while also beginning to explore what inclusive AI looks like in practice. It means involving disabled people in the design and testing of new interaction models. And it means recognising that accessibility is not tied to a particular technology, but to the outcomes we create for users.

At AbilityNet, we see this as a natural extension of our work. The tools may change, but the goal remains the same: ensuring that digital technology works for everyone.

Looking ahead - and acting now

The shift towards conversational, agent-driven experiences will not happen overnight. There will be a long period where traditional interfaces and AI coexist, overlap, and evolve together. That gives us a valuable window to shape what comes next.

The organisations that succeed will be those that think holistically - not just about compliance, but about experience; not just about technology, but about people. And right now, the most immediate priority remains clear. Your websites and apps still need to be accessible, usable, and inclusive for the widest possible audience. They are the foundation on which everything else will be built.

If you want to ensure your digital services are ready both today and tomorrow, AbilityNet can help. Our accessibility audits, diverse user testing, and strategic support can help you create inclusive experiences now, while preparing for the changes that are already on the horizon.

More from Robin Christopherson

Explore more of Robin's thinking on AI, accessibility, trust, governance and inclusion.

AI could make financial services more inclusive. But only if accessibility is part of the infrastructure

Published 27 July 2026. Why accessibility must be treated as core infrastructure if AI is to improve financial inclusion.

Three stories about AI and accessibility that every CTO will recognise

Published 21 July 2026. Real-world examples showing why accessibility is becoming a measure of AI trust, governance and user experience.

AI skills are becoming critical infrastructure. Not optional learning.

Published 6 July 2026. Why AI literacy is becoming essential and what organisations must do to ensure inclusive outcomes.

The trust gap: why AI's edge cases matter to customer loyalty

Published 30 June 2026. Why accessibility and edge cases are critical to building trust in AI-powered services.

The delicious irony at the heart of trusting AI

Published 17 June 2026. A personal reflection on trust, accessibility and the unexpected ways AI can both help and hinder.

Executive briefings for senior leaders

Robin offers invite-only executive briefings exploring AI, accessibility, governance, customer trust, regulation and digital inclusion.

Contact us to discuss how a briefing can help give accessibility a place in your AI roadmap

Let us help you on your AI journey. See our range of AI integration services and contact us for more information.