Summary
On the most problematic site, agents completed 31% of assigned tasks compared with 96% after accessibility fixes. The study also reported up to six times higher token use on inaccessible sites.
Why it matters
The result links accessible HTML and interface semantics directly to agent usability, making accessibility an emerging reliability and efficiency requirement for agent-facing websites.
Related coverage
- AutonomyAI connects product discovery directly to review-ready pull requests
- Submersion AI launches Basin cybersecurity reasoning model
- Gurucul releases runtime detection and prevention for risky AI behaviour