3.5: AI and Accessibility

AI tools have the potential to expand access to learning—especially for students with disabilities—by generating alternative formats, offering personalized support, and enhancing communication. However, without intentional design and oversight, AI may also create new barriers or reinforce exclusion.

⚠️ When AI Falls Short on Accessibility
AI can inadvertently exclude users if not designed with diverse abilities in mind:

• Inaccessible outputs: AI-generated content (e.g., images, charts, or videos) may lack alternative text, captions, or screen reader compatibility.
• Inconsistent language or formatting: Automated responses may use jargon, inconsistent headings, or confusing structure that impedes comprehension for neurodivergent or visually impaired users.
• Voice, text, and translation errors: Voice-to-text tools may misinterpret non-standard speech patterns, while translation models may fail to preserve meaning for multilingual learners or users with AAC devices.

✅ Using AI to Enhance Accessibility
When used intentionally, AI can support Universal Design for Learning (UDL) and inclusive teaching practices:

• Generate alt-text for images, closed captions for video content, or plain-language summaries
• Translate materials into multiple languages or convert written content into audio
• Adjust reading levels or simplify complex texts for cognitive accessibility
• Design accessible layouts using clear headings, lists, and consistent formatting

---

---