GenAI and Ethics

• Dilemma: GenAI tools can be used by students to generate content or complete tasks independently, raising questions about originality, authorship, and the nature of student learning.
• Consideration: Instructors should consider how to balance GenAI’s role as an aid while preserving the authenticity of student work. Setting clear guidelines for when and how GenAI tools can be used (such as using it only for brainstorming or drafting) can help, as does designing assignments that require personal insights or reflections AI cannot provide. In addition, engaging students in discussions around core values in academia like transparency, fairness, privacy, and academic integrity will promote ethical use of these tools.

• Dilemma: GenAI models often reflect biases present in the data they are trained on, which can lead to unequal representation or perpetuation of stereotypes.
• Consideration: Instructors should be mindful that AI outputs may not represent all perspectives fairly and can reinforce bias. When using AI-generated materials, instructors can critically assess and supplement them with diverse perspectives to ensure an inclusive learning environment.

• Dilemma: Using GenAI in class without informing students could erode trust or lead students to misunderstand the source of instructional content and feedback.
• Consideration: Transparency is crucial, especially when GenAI is used for tasks that directly impact students, like grading or content delivery. Instructors should communicate how GenAI is being used, allow for student input or consent when appropriate, and clarify that it is a supportive, not standalone, teaching tool.

• Dilemma: Many GenAI tools require inputting data, which may include sensitive or identifiable student information, raising concerns about data privacy and security.
• Consideration: Instructors should ensure compliance with institutional privacy policies and avoid using AI tools that compromise student privacy. Choosing GenAI tools with strong privacy protections and limiting input to non-identifiable information can help protect the integrity of student data.

• Dilemma: If students and teachers become overly reliant on GenAI, there is a risk that essential skills, such as critical thinking, writing, and research, may be undermined.
• Consideration: Instructors can encourage responsible use by framing GenAI as a tool to enhance—not replace—learning. Emphasizing foundational skills and fostering critical engagement with AI outputs (e.g., fact-checking, analyzing biases) can support skill development.

• Dilemma: Not all students have equal access to GenAI tools, which may create disparities in learning opportunities and performance if these tools become essential.
• Consideration: Instructors should be mindful of students’ access to AI tools. Where possible, provide institutional access to necessary AI tools or alternative resources so all students can participate equitably.

• Dilemma: GenAI tools can produce content that is derivative or synthesized from copyrighted sources, including art, music, photos, and published books. It can also misappropriate Indigenous content. All of this raises questions about intellectual property rights, perpetuates cultural misappropriation, and erodes trust.
• Consideration: Instructors should be aware of the intellectual property implications of using AI-generated content. In addition, it is important to consider the need for respect, transparency, and accountability, ensuring that AI outputs do not commodify or distort culturally significant Indigenous knowledge, but instead foster genuine understanding and equitable collaboration with Indigenous communities. Encouraging students to create their own work and requiring citation of AI sources when used will identify original work and help avoid potential copyright infringements and promote respect for original creators.

• Dilemma: AI-generated content can be prone to “hallucinations” (confidently incorrect information) or oversimplifications that may mislead students or compromise learning quality.
• Consideration: Instructors should fact-check AI-generated content and encourage students to do the same, fostering a habit of critical evaluation. Instructors may also highlight that AI is not infallible and teach students how to recognize and cross-reference accurate information.

• Dilemma: Integrating GenAI could change traditional teaching methods, potentially leading to a reduction in teacher-student interactions and shifts in pedagogical focus.
• Consideration: Teachers should strive to maintain meaningful interactions with students and emphasize the human aspects of learning, such as mentorship and personalized feedback, ensuring AI complements rather than reshapes the teaching process.

• Dilemma: Running GenAI models often requires significant energy and computing power, contributing to carbon emissions and environmental strain. As well, data workers and content managers, many of whom are based in the Global South, often face exploitative working conditions with their contributions largely overlooked in the push for innovation.
• Consideration: Instructors can consider the environmental impact of using GenAI tools by reducing the frequency of use and leveraging these tools thoughtfully while advocating for fair labour practices. Educating students about the environmental cost of AI as well as values of equity, respect, and transparency toward those who make these innovations possible can also encourage mindful use.

Ethical AI for Teaching and Learning (Cornell University: Center for Teaching Innovation)

Ethical Concerns in AI Higher Education (University College Dublin Pressbooks Network)