Skip to main content
ArthurAI
  • For educators
  • How it works
  • Dawn Directive
  • Insights
  • Technology
  • Trust
Talk to our team →
Editions
  • SLEK-12 districts
  • ULEhigher education
  • VLEworkforce + training
  • CLEL&D and compliance
  • PLEComing soon
More
  • For educatorsTeachers, faculty, instructors, trainers
  • How it worksTen capability deep-dives, all code-true
  • Dawn DirectiveCIAI’s 12-month AI fluency certification
  • InsightsIssue 01 — editorial from the founder
  • TechnologyEve-Education™ F5/reasoner
  • TrustCompliance, security, data handling
Sign in
  • Sign in to SLE →
  • Sign in to ULE →
  • Sign in to VLE →
  • Sign in to CLE →
Talk to our team →
Trust CenterAccessibility conformance

WCAG 2.1 AA. From day one.

Accessibility is a launch criterion at ArthurAI™, not a roadmap item. The standard we target is WCAG 2.1 AA — the standard cited by Section 508 of the Rehabilitation Act and by the Title II ADA Web Rule (28 CFR Part 35). Public-sector and many institutional buyers require demonstrated WCAG conformance; the platform is built to satisfy that requirement on day one of any deployment.

What WCAG 2.1 AA means in our build

  • Perceivable. Color contrast meets the 4.5:1 ratio for body text and 3:1 for large text. No information is conveyed by color alone. Text alternatives exist for non-text content. Educator-uploaded media prompts the educator for alt text at upload time.
  • Operable. Every interactive surface is keyboard-navigable. Focus is always visible. Skip-links jump past the navigation. No keyboard traps. Custom UI components carry appropriate ARIA roles and states.
  • Understandable. Form labels are present and correctly associated. Error messages are programmatically linked to the field. Language of the page is set and updates correctly when locale switches. Predictable navigation across pages.
  • Robust. Semantic HTML structure (landmarks, headings, lists). Custom widgets follow the WAI-ARIA Authoring Practices. The platform tests against current screen-reader behavior (NVDA, JAWS, VoiceOver, TalkBack).

How we verify

  • Automated checks at build time. Every product portal ships with axe-core integrated into the test suite. The CI pipeline blocks merges that introduce accessibility regressions detected by axe-core.
  • Lint-time enforcement. eslint-plugin-jsx-a11y flags accessibility anti-patterns at the source-code level (missing alt text, click handlers without keyboard handlers, missing form labels, etc.).
  • Manual review per release. Each release is reviewed for keyboard navigation, screen-reader behavior, color contrast in branded themes, focus management, and text alternatives across every learning-flow surface.
  • RTL-aware design. Document direction flips automatically for Arabic and Urdu locales. Layouts are mirrored, not just translated.

Section 504 and Section 508 alignment

Section 504 of the Rehabilitation Act and Title II of the Americans with Disabilities Act place obligations on the educational institution; ArthurAI™ supports those obligations with accessible interfaces and content. Section 508 of the Rehabilitation Act applies to federal procurement; the WCAG 2.1 AA target satisfies the technical baseline cited by Section 508.

VPAT (Voluntary Product Accessibility Template)

A formal VPAT 2.5 document covering WCAG 2.1, Section 508 Refresh, and EN 301 549 (European accessibility standard) is available to institutional buyers under NDA. The VPAT is updated on every major release that touches user-facing surfaces. To request the current VPAT, talk to our team.

What we do not claim

We do not claim WCAG 2.1 AAA conformance. AAA is a stricter standard not applicable to all content, and not required by Section 508 or Title II. We target AA, verify AA, and represent AA. We also do not claim conformance for third-party content educators upload to the platform; we provide tools that help (alt-text prompts at upload, automated contrast checks on themed backgrounds), but the educator is responsible for the accessibility of content they author.

See also: accessibility statement · full compliance posture · data handling.

End of issue
ArthurAI

The Agentic Learning Operating System. Reasoning-first. Educator-attested. Four editions in market.

SLE
  • SLE overview
  • Capabilities
  • Safety & disclosure
  • For teachers
ULE
  • ULE overview
  • Capabilities
  • Faculty of record
  • For faculty
VLE
  • VLE overview
  • Capabilities
  • For instructors
CLE
  • CLE overview
  • Capabilities
  • SOC 2 readiness
  • For trainers
Technology
  • Technology overview
  • F5/reasoner
  • Eve-Genesis (Education)
  • Eve-Grid
  • Multi-tenant isolation
  • Multi-language
  • Integrations
Trust
  • Trust Center
  • Compliance
  • Security
  • Security FAQ
  • Subprocessors
  • Data handling
  • Disclosures
  • Responsible AI
  • Accessibility conformance
  • DPA
  • Incident response
Company
  • Why ArthurAI
  • About
  • Founding story
  • Three regions
  • Leadership
  • Deployments
  • Pricing
  • Contact
  • MindHYVE.ai
  • Privacy
  • Terms
  • Accessibility
PLE
  • PLE — Coming soon
For educators
  • For educators
  • Your day with SLE
  • Your semester with ULE
  • Your program with VLE
  • Your quarter with CLE
  • Your first month
Resources
  • Insights — Issue 01
  • The Dawn Directive
  • Resources
  • Whitepapers
  • Press
  • Changelog
MindHYVE Product Family
Agentic Operating Systems
  • ArthurAI™ (this site)
  • ChironAI™ →
  • JustineAI™ →
  • TheoAI™ →
Research surfaces
  • IPSC™ →

One parent, one architecture, four shipped Agentic OSes plus a research corpus. See the family in detail →

Newport Beach · Islamabad · Nairobi© 2026 Eve-Education, LLC, a wholly-owned subsidiary of MindHYVE.ai, Inc.. All rights reserved.

ArthurAI™, MindHYVE.ai™, Eve-Education™, Eve-Fusion™, Eve-Grid™, Eve-Genesis™, HYVE™, HYVE Labs™ are trademarks of MindHYVE.ai, Inc.. ChironAI™, JustineAI™, and TheoAI™ are trademarks of MindHYVE.ai, Inc.. The Dawn Directive™ is a trademark of the California Institute of Artificial Intelligence (CIAI), used under license.

Powered byMindHYVE.ai