Generative AI in Higher Education (BMC)

Recommended strategies involve leaning on active learning, transparent teaching, and UDL strategies that support inclusive engagement and success:

• Clearly communicate whether and how AI tools can be used in each course.
• Clearly communicate the reasons behind your policy choices and how they relate course learning goals. 
• Incorporate metacognitive elements into assignments (i.e., asking students to reflect on their learning, discussions, research processes, etc.)
• Use authentic assessments and/or have students demonstrate learning in multiple ways.
• Invite students to actively and critically engage with AI generation tools or the issues they raise.  

Sources generally caution against:

• Redesigning assignments to take advantage of current AI limitations (e.g., requiring citations, using archival sources and articles behind paywalls). AI technologies and economics evolve so rapidly that windows of success will be limited.  
• Relying on "AI-detection" tools. Not all purveyors are reputable and researchers have found considerable risk of false positives, especially when analyzing work by novice writers or writers for whom English is a second language.  

These articles suggest assignments and best practices for incorporating the study of generative AI tools and output into courses.   

There is no monolithic "student perspective" on generative AI: interests in and concerns about the technology and its applications vary among students as they do among faculty. However, students are more likely to believe that AI will shape future work and societies, and that they will need to engage with AI at some point in their lives.