7 final principles for the use of Generative AI
in Higher Education
The list of principles below is the final result of the Students' Assembly. Since the assembly, these principles have been reworked to correct errors of syntax or grammars. Out of the 9 principles originally drafted, 7 remain (4 principles have been merged into 2 upon the request of assembly members).
1) We value experimentation.
We believe in embracing emerging technologies, including generative AI (GenAI), and their capacity to enhance learning and development. We should not be afraid to experiment, unlock our inner creativity and use these tools to foster both imagination and innovation. We value playfulness in teaching and learning. We should feel inspired.
2) We embrace sustainability-driven innovation and conscious change.
We believe that it is our responsibility to be conscious and proactive of the environmental impacts of our AI uses. We must question the appropriateness and scope of the GenAI tools based on a cost-benefit analysis of their environmental footprint. Is the task at hand worthwhile, considering its ecological impact? If so, go create great things. If not, use other, more suitable and sustainable tools. As importantly, we believe that the university should openly communicate on the ecological consequences of the use of GenAI.
3) We believe in upholding the integrity of education.
GenAI should be used as a tool to enhance, not replace learning and teaching experiences. We value assessments and learning opportunities which emphasize engaged, thoughtful, critical, and creative contributions through multiple approaches, which may or may not include the use of GenAI tools. Intentionality is needed wherever these tools are encouraged or restricted.
4) We advocate for an empathetic understanding of individuality.
We respect individual approaches to the use of GenAI, and we believe in the right to refusal in the use of GenAI for both students and teachers to be respected. The choice of using or not using GenAI is left up to the individual, with the understanding that it serves to further learning and teaching and is not detrimental to the integrity of EFI. This means that students and teachers can refuse to be judged for using GenAI and refuse to use it in their work in so far as this use respects the principles hereby mentioned.
We strive to be conscious of the impact of GenAI use. We will be mindful of individuals' rights, needs, and desires both at EFI and within society. We acknowledge the possibility of human and AI error and the biases and human cost that may be present. We support improvements in the broader social contexts and aim for positive future developments in this area.
5) We strive towards trust and transparency.
We should be able to be held accountable for the way in which we use GenAI. This means we should be able to verbalize why and how we’ve used GenAI. We should take responsibility for its use and how it is implemented in assignments and projects; recognizing that it is a tool, not a secret to be kept. We should honor the principles of how to use genAI by using clear guidelines/rules that foster the implementations of these principles. We need to be honest, genuine and respectful.
In a similar manner, we expect EFI to overtly disclose all uses of GenAI (e.g., in communications, feedback, marking, teaching materials, or other organizational tasks). The institute should be open about the investment choices and procurement processes underway, as well as the third party tech providers involved. In the spirit of transparency, we believe in open source technologies and democratically-governed technologies.
6) We believe in a democratically governed GenAI.
We encourage active participation and open lines of dialogue between the institution and students during any regulation drafting about GenAI and throughout the implementation phase. We believe in the active role of student representatives in co-designing new regulatory measures or guidelines regarding GenAI.
We believe that the engagement of students allows for the development of an AI governance grounded in community concerns, narratives and social justice. This principle is presented in the spirit of encouraging an institutionalized culture of deliberation and collaboration throughout all levels of EFI.
7) We strive towards equality, inclusion and accessibility.
GenAI must not be a tool that excludes individuals or creates further inequalities in EFI. Democratizing access to GenAI to support a variety of learning needs is crucial. GenAIs that are introduced in the learning experience should have user-friendly interfaces to accommodate accessibility needs, including those of students, staff or academics with disabilities. We aim to foster an inclusive and just environment where GenAI serves as a catalyst for learning.