Using AI in Scaling Accessibility 

As we work towards meeting digital compliance deadlines, the sheer volume of course materials requiring updates can feel overwhelming. Artificial intelligence offers a potential way to accelerate this work, acting as an assistant for some time intensive tasks. However, using these tools effectively requires a balanced approach that pairs machine efficiency with human judgment. 

Large language models can quickly generate initial drafts of alternative text for complex images or suggest logical heading structures for long, unformatted documents. For video content, AI produced transcription provides a good starting point for captions which can significantly reduce the time needed to make lectures accessible to all students. 

Despite these strengths, AI is not a perfect solution. Automated tools can fail to correctly interpret nuance, context, and specialized academic terminology. AI might identify a “tree” in a biology diagram but miss the specific botanical structures essential for the lesson. Similarly, auto-generated captions often struggle with technical terms. Without human review, these automated fixes can inadvertently introduce new barriers or misinformation. 

The proper role of AI is to serve as a bridge to the final destination. By viewing AI as a partner in the process, we can meet our accessibility goals without sacrificing the quality of our instructional materials. 

How are you currently integrating AI tools to help streamline your course design or accessibility workflows?