By LINDSAY ONUFER

Conversations about how to create accessible learning experiences in higher education often begin and end with providing disabled students (see note below) with the reasonable accommodations they are entitled to under the Americans with Disabilities Act.

While providing accommodations is a crucially important, ethically and legally necessary step toward accessibility, accommodations alone cannot remove all the barriers that students face when they enroll at a university. In “Academic Ableism” (2017), disability studies scholar Jay Dolmage described these literal and symbolic barriers as “steep steps” that perpetuate inequity and exclusion:

"University campuses have lots of steep steps-but the entire university experience can also be metaphorized as a movement up steep steps. The steep steps, physically and figuratively, lead to the ivory tower. The tower is built upon ideals and standards —historically, this is an identity that the university has embraced. I want to suggest that we have mapped the university in this way — as a climb up the stairs of the ivory tower — for particular reasons. Often, maps are created not to reveal exclusive, but to create it."

Dolmage argued that accommodations help but, like a ramp installed next to steep steps, are implemented as “retrofit” fixes for an inherently ableist system. Beyond retrofitting accommodations, designing for accessibility from the outset can help create more inclusive learning experiences and mitigate some of the obstacles and challenges that disabled students might otherwise face.

Many instructors are motivated to adopt more inclusive teaching approaches. However, they often find it challenging and time-consuming to identify, create and implement new resources and strategies.

Generative artificial intelligence can help. Artificial intelligence (AI) tools, including generative AI tools like ChatGPT, have been hailed as potentially transformative for improving accessibility. For many people with disabilities, the promise of these tools has yet to be fully realized in part because AI is often not designed specifically to improve accessibility or with disabled users in mind (Smith & Smith, 2020).

However, as students and higher education institutions embrace the use of generative AI to enhance teaching and learning, we should also begin exploring how these tools could be used to help instructors enhance course accessibility and inclusion. Below, I have listed some suggestions for how to get started.  

Implementing Universal Design for Learning

Universal Design for Learning (UDL), a framework inspired by the universal design of architecture and grounded in decades of cognitive science research, is widely advocated to improve inclusivity and supplement individualized reasonable accommodations. UDL Guidelines propose that giving students choices and varied means to engage with different types of course content and express their learning improves learning experiences for all students, and students with disabilities in particular. You can input your course materials or assessments into generative AI tools like ChatGPT or compose prompts to solicit suggestions to help you create or revise the materials to incorporate UDL guidelines.

Here are some sample prompts:

Act as a higher education inclusion and accessibility expert. Offer suggestions to revise the course materials, learning task or assessment to incorporate Universal Design for Learning guidelines by:

• Examining and proposing revisions to my course policies to make them more inclusive  

• Giving students choices about how to express their learning

• Explaining and emphasizing the relevance and value of the content or assessment

• Promoting intrinsic motivation

• Developing self-assessment and reflective skills

• Fostering students’ learning self-efficacy

• Explaining concept in a different way

Although generative AI does not always provide flawless output, using it as a brainstorming tool or virtual sounding board gives you a starting point and makes planning course revisions quicker and easier.

Encouraging students to use AI for learning support

In addition to using AI to makeover your teaching materials, you can give students specific suggestions for using generative AI to support their learning, which can be especially beneficial to neurodiverse students and students with cognitive or learning disabilities.

Before instructing students to use AI, faculty should determine and explain which types of uses would be acceptable and which would constitute academic integrity violations to students. The Center for Teaching and Learning recommends that instructors use an AI syllabus statement to clarify expectations. Preventing AI misuse requires teaching students how to use these tools ethically and effectively. 

Some examples might include telling students to use generative AI tools to:

• Act as a “study buddy” and quiz the student on course concepts or create a study guide in preparation for an exam or quiz

• Simplify the language of course assessment instructions to make them easier for students to understand

• Explain course concepts in multiple ways

• Generate examples or practice problems

• Help students organize their notes or ideas

• Create a step-by-step plan for how to tackle an assessment or to use your feedback to improve their learning

As Dolmage emphasized, disabled students face a myriad of challenges in universities. While accommodations provide necessary support to navigate these “steep steps,” Generative AI can help faculty more efficiently and proactively design more accessible learning experiences.  These suggestions are not comprehensive but will hopefully inspire you to explore how Generative AI tools can support creating more inclusive courses.

Lindsay Onufer, is program manager and senior teaching consultant in the University Center for Teaching and Learning.

NOTE FROM THE WRITER: I used person-first language (“students with disabilities”) and identity-first language (“disabled students”) interchangeably in this article per National Institutes of Health style recommendations. Although person-first language is more commonly used in the United States, some disabled people and communities have rejected person-first language. When writing or speaking about or to an individual with a disability, best practice is to use the language that individual prefers.    

TO LEARN MORE:

• Read “How ChatGPT Could Help or Hurt Students with Disabilities” in Faculty Focus

• Review the University Center for Teaching and Learning’s Generative AI Resource for Faculty website

• Register for Pitt’s 2024 Assessment and Teaching Conference, which will feature an undergraduate student panel discussing generative AI, moderated by Annette Vee, associate professor and Composition Program director

• Register for one of the Teaching Center’s spring workshops on anti-ableist pedagogy, Universal Design for Learning, or digital accessibility.

• Using GPT-4? Watch this short video to learn how to use it to generate alt text image descriptions.  

REFERENCES

CAST’s About Universal Design for Learning

James T. Dolmage, (2017). “Academic ableism: Disability and higher education,” University of Michigan Press.

Peter Smith and Laura Smith (2021). “Artificial intelligence and disability: Too much promise, yet too little substance?” AI and Ethics