Our Research Direction: Designing for Accessibility
In our initial research, we found that a significant barrier to digital equity was the “accessibility gap” – the time between releasing a new feature and creating a support layer for it. To fill this gap, we move from reactive tools to interface-native agent systems.
Research Pillar: Using Multi-System Agents to Improve Accessibility
Multimodal AI tools represent one of the most promising avenues for creating accessible interfaces. In specific prototypes, such as our work on web readability, we have tested a model in which a central orchestrator acts as a strategic reading manager.
Instead of a user navigating a complex maze of menus, Orchestrator maintains a shared context: it understands the document and makes it more accessible by delegating tasks to expert subagents.
- The synthesis agent: masters complex documents by breaking down information and delegating key tasks to expert sub-agents, making the most in-depth information clear and accessible.
- The Settings agent: Handles UI adjustments, such as text scaling, dynamically.
By testing this modular approach, our research shows that users can interact with systems more intuitively, ensuring that specialized tasks are always handled by the right expert without the user needing to search for the “correct” button.
Towards Multimodal Fluidity
Our research also aims to move beyond basic speech synthesis toward multimodal mastery. Leveraging Gemini’s ability to process voice, vision, and text simultaneously, we built prototypes capable of transforming live video into immediate interactive audio descriptions.
It’s not just about describing a scene; it’s about situational awareness. In our co-design sessions, we observed how allowing users to interrogate their environment interactively – requesting specific visual details as they occur – can reduce cognitive load and transform a passive experience into an active, conversational exploration.
For more information on how AI agents can redefine universal design to increase accessibility, visit the source link Here.
“`
