KI für axe vorantreiben: Der nächste Sprung in der digitalen Barrierefreiheit

Deque

Von „ Deque -Systeme“

25. Februar 2025

Blog-Kopfzeile: Weiterentwicklung der KI für Axe

At Deque, we are working diligently toward a world where accessibility testing, fixing, and development require zero specialized knowledge—where any team, anywhere, can build accessible experiences effortlessly.

Artificial intelligence is a crucial factor in reaching this goal.

Today, automated testing covers 57% of accessibility issues by volume. This represents incredible progress. But we’re pushing this even further. At axe-con 2025, we are announcing new, AI-driven capabilities that address the toughest accessibility challenges—increasing test coverage, reducing manual work, and making accessibility testing faster and easier than ever.

New capabilities coming in 2025 include advanced rules, AI-assisted testing for Intelligent Guided Tests (IGTs), accessible coding with AI agents, and Axe Assistant for Slack and Teams.

Advanced rules: Smarter automation

We’re building on axe-core with a new ruleset powered by AI, advanced machine vision, and screenshots to detect accessibility issues commonly missed by traditional automation—all while upholding our commitment to zero false positives.

Automation, for example, can detect when an image lacks an alt attribute, but it can’t always determine if an empty alt is correct (e.g., a decorative icon) or an issue (e.g., a missing label on a button). Developers currently have to make that call manually.

By analyzing real-world data from thousands of audits, we’re closing these kinds of key gaps. Our initial release will increase automated coverage by up to 10% by volume—a major leap forward in accessibility test automation.

These advanced rules will first be available in the axe DevTools Extension.

To learn more, register for “AI Systems for Accessibility” at axe-con.

AI-assisted testing: Automating Intelligent Guided Tests

Intelligent Guided Tests (IGTs) can deliver 80% coverage by volume today, but they require costly manual effort—developers have to answer questions to help AI understand their application.

With our new testing approach, AI can run these IGTs automatically—analyzing pages, applying reasoning, answering questions, and delivering results with clear explanations. Developers can then accept, reject, or refine AI suggestions.

AI also makes re-testing faster by remembering past decisions. During regression testing, it flags only what has changed—reducing AI workload and keeping accessibility checks efficient.

To learn more, register for “Automation Gains: Increasing what can be automated in accessibility testing” at axe-con.

Axe Assistant: AI-powered guidance, where you work

Educating your teams is the best way to fix accessibility issues before they reach your customers. Coming soon to Slack and Teams, axe Assistant gives designers, testers, and developers direct access to Deque University’s trusted knowledge base, offering expert guidance without disrupting their workflow.

Axe Assistant provides real-time accessibility insights, whether working solo or collaborating in group chats, helping teams catch issues early. Developers can quickly check color contrast, generate accessible code, and simplify complex guidelines into plain language—all without leaving their workspace.

By embedding axe Assistant into the tools teams already use, we’re making accessibility easier, more intuitive, and always within reach.

Watch the following video to see axe Assistant for Slack in action.

Accessible coding with AI agents

AI coding agents are becoming a key part of the development workflow, allowing developers to generate, refine, and validate code with minimal manual effort. These agents work iteratively, taking instructions, producing results, gathering feedback, and improving their output until it meets the developer’s needs.

By connecting axe Assistant and axe Tools to AI coding agents, accessibility becomes an integral part of this process. Deque is the first accessibility company to support the emerging standard for integrating tools into AI agents—the model context protocol.

The agent will be able to pull in accessibility expertise whenever needed—asking axe Assistant for accessibility requirements, generating accessible code, and using axe Tools to validate and fix issues automatically. All of this happens seamlessly within the developer’s workflow, with the AI agent handling the heavy lifting while the developer stays in control.

The result? Faster, more accessible code without guesswork. Developers can focus on building great experiences, knowing that accessibility is built in from the start.

To learn more, register for “The future of the axe Platform” at axe-con.

Join us in shaping the future

We’re inviting our current customers to join our early access program for advanced rules, AI-assisted testing for IGTs, accessible coding with AI agents, and axe Assistant for Slack and Teams. Ready to be part of the next evolution in accessibility testing? Request access today!

If you haven’t already, register for axe-con right now to learn more about the new capabilities from the Deque team.

Deque

Deque

Deque is the global leader in digital accessibility, helping the world’s top enterprises build inclusive products, services, and experiences and achieve lasting compliance. Recognized by leading industry analysts for its AI-powered tools, comprehensive services, and developer-trusted solutions, Deque delivers the industry’s most complete accessibility offering. The Axe platform, anchored by Axe-core, has more than 5 billion downloads and 800,000 installed extensions, making it the global standard for accessibility testing. As a pioneer of people-first accessibility, Deque applies a human-in-the-loop approach that blends expert insight with AI innovation to advance its mission of digital equality for all.

Erhalten Sie Blog-Beiträge direkt in Ihren Posteingang

Kein Geschwafel, sondern echte Erkenntnisse zum Thema Barrierefreiheit von qualifizierten Experten.

Sie erklären sich damit einverstanden, dass Deque Informationen gemäß den Bestimmungen in DequeDatenschutzerklärungbeschrieben, Informationen von Deque entgegennimmt, nutzt und weitergibt. Sie können Ihre Einwilligung jederzeit widerrufen, indem Sie uns kontaktieren.

Mehr zu diesem Thema

Inside the AI software factory: Where accessibility is won or lost

Dylan Barrell
November 19, 2026 By Dylan Barrell

As your organization moves development into AI software factories, you must fully embed digital accessibility, just as you must integrate privacy and security. Get it right, and your products, services, and experiences become and stay accessible. Get it wrong, and you increase risk, cost, and technical debt.

Artikel lesen
5 areas where your software factory must get accessibility right. Clear accessibility requirements. Deterministic checks. Well-steered agents. Independent proof that the work was done. Monitoring after release.

How to choose the right VPAT approach for your organization (and whether you should use AI to complete your ACR!)

Will Creedle
November 4, 2026 By Will Creedle

In this post, I’m going to discuss why accuracy matters for digital accessibility in an era of evolving global standards and increased regulatory scrutiny. I’ll explain where and how VPATs and ACRs fit into your accessibility strategies, and what you need to know before making any decisions about AI use.

Artikel lesen
Accessibility professionals working at a laptop in an office. Image is overlaid with four text boxes containing the terms: Conformance reporting, VPAT, artificial intelligence, compliance strategy.