Zoek in de site...

Guidelines AI usage

Developments in the field of generative artificial intelligence (GenAI) are progressing rapidly, and naturally, there are many questions about its impact on education. GenAI has much to offer in education, but it also has intrinsic shortcomings, can hinder skill acquisition, and raises questions about ethics, privacy, safety, and academic integrity. To provide guidance for students and teachers, the Faculty of Arts has established the following guidelines regarding the use of GenAI in education.

What do we mean by GenAI?

GenAI refers to applications that can generate text, images, music, and other content (output) based on a written or spoken instruction (prompt). GenAI applications use models that have learned to recognize patterns in large amounts of training data. With this information, content can be generated based on statistical probability.

Things to be aware of

  • GenAI applications do not understand the world; they only perform probability calculations. While it's not ruled out that they may create accurate content, they often produce content that does not align with reality, a phenomenon referred to as "hallucination."
  • For very specific, specialized, recent, or local topics, GenAI applications may not be able to generate accurate output, as these topics are likely underrepresented in the training data.
  • Often, it's unknown which data was used to train the models on which GenAI applications are based. This is problematic, as the output of GenAI applications is only as reliable, representative, and fair as the training set. Information in the training set may be incorrect, outdated, and may contain biases related to demographics, gender, and beliefs.
  • It's not always clear what developers of GenAI applications do with the texts you input as prompts.

Guidelines

  • Do not present the output of GenAI applications as your own work. Products produced as part of your studies must unequivocally demonstrate that you have acquired knowledge and skills as described in the learning objectives of the course in which your product plays a role. Presenting GenAI output as your own work is considered fraudulent.
  • Mention the use of GenAI and refer to it correctly. We want students to understand what GenAI is, what you can and cannot do with it, and how you can research it. For this reason, you may receive permission or explicit instructions from a teacher to use GenAI. Always describe in a first footnote, attachment, or postscript in general terms how you used GenAI, including the prompts used, and use the referencing style used in your study program to refer to AI generated text and images. It is the responsibility of the teacher to make choices and provide instructions on this. You can familiarize yourself with current referencing practices for GenAI on the following websites:
  • Compare the output of GenAI applications with other scientific sources. As mentioned, the output of GenAI applications may contain inaccuracies and biases. Be critical of the output and use additional scientific sources to assess the quality of GenAI output.
  • Do not use personal data or intellectual property in your prompts. Since you don't know what happens to the information in your prompts, sharing this data may violate the General Data Protection Regulation (GDPR) or intellectual property rights. You can only do this if Radboud University has entered into a processing agreement with the supplier of the GenAI application you use, which regulates the protection of personal data and sets out the appropriate security measures.
  • A teacher cannot compel you to create an account for GenAI applications for which Radboud University does not have a license. If teachers want to use GenAI in their teaching, they must consider providing an alternative if you do not want or cannot create an account for an external GenAI application. Teachers can do this, for example, by generating the output themselves for you to analyze or by choosing a learning activity where they collectively input prompts into a GenAI application suggested by you and your classmates.

When is the use of GenAI allowed?

  • You may use GenAI as a search engine to guide you to relevant publications. Be critical: GenAI applications can also hallucinate publications. Never refer to a publication you have not seen yourself.
  • You may use GenAI to learn new material or practice with it. For example, ask GenAI applications for explanations of concepts you struggle with, examples, or practice questions. Again, be critical and check the quality of the GenAI output. As mentioned, GenAI applications likely cannot generate accurate output for very specific, specialized, recent, or local topics, as these topics are presumably underrepresented in the training data.
  • You may use GenAI to generate templates for inspiration for products you are going to create. For instance, ask for fixed elements, the structure, or typical style features of an essay, research report, advisory report, table, graph, diagram, presentation, infographic, etc.
  • You may use GenAI to revise or improve products without generating new content. For example, you can use GenAI to check your texts for grammar, spelling, punctuation, deviations in sentence length, or noticeable stylistic breaks. For an optimal learning effect, do not use prompts asking GenAI to improve your text; instead, ask for a list of improvement suggestions in specific areas and critically evaluate them before incorporating them. Note: This does not apply to assignments where the ability to revise or improve texts is an explicit learning objective. For instance, if the goal of an assignment is to teach you to rewrite a text for a different audience, check a non-native text for grammatical errors, or assess whether a text adheres to a correct scientific style, then you must do it yourself.
  • You may use GenAI to create content if you have permission from your teacher. This could be an assignment where you critically reflect on the accuracy, relevance, or quality of content produced by GenAI, an analysis of the biases present in the output of specific GenAI applications, or a project in which you learn to write better prompts and effectively use GenAI. Always follow the guidelines when creating content and consult your teacher if in doubt.