Phil Leonard is a Professor in the Department of Humanities.

Phil described two assessment activities in which he has asked students to interact critically and reflectively with Generative AI (GenAI) tools on the same module. In the two years he has run these activities, he observed a stark difference in student prior knowledge of the tools.

How did Phil use AI in his practice?

In a new first-year module about how technologies are changing literature (and how we study it), Phil designed and implemented two assessments in which students were asked to engage with GenAI tools:

• Designing a book cover, including the front and back, and other elements of book covers such as blurb, reviewer endorsements, and barcode. Students were asked to use GenAI tools to produce the visuals and submit a commentary justifying the design and a reflection on the process alongside their book cover.
• Generating an essay using GenAI, which is then used as a source for a reflective essay on how well the generated essay fulfils its brief, or even to what extent it can be considered an essay.

These activities are an introduction to the creative and ethical dimensions of GenAI, incorporating theory and philosophy as far as possible for a first-year module.

What made Phil decide to use GenAI within this module?

The subject matter of this module includes a history of artificial writing systems, and of course this now includes GenAI. Students on this module learn to situate these new tools in a larger historical narrative stretching back to the machine generated poetry of the 1960s, and to earlier writing technologies such as pencil and paper. Students looked at examples of news stories about GenAI relating to artists and authors in order to think about how this latest evolution in writing technology is impacting literature and literary studies. Phil wants his students to recognise that technological innovations don’t always lead to older technologies being abandoned, and it can be reductive to pit GenAI technologies against ideas of authenticity.

But also from a skills perspective, Phil can see how the popularisation of GenAI writing tools is going to affect students’ working lives, and he’s noticed that his students have been thinking about this too. Phil sees their engagement with this technology as absolutely essential to prepare them for their future employment but also wants them to develop a nuanced view of these tools through encountering them in the mediated space of this module.

What was the impact of using GenAI for this activity on the students?

Phil’s students were able to creatively envision and produce book covers, which is an engagement with literary culture that would likely be unavailable to them within this module without the use of GenAI technology. He was impressed with the ingenuity of some of the submitted work.

Through the essay assessment students were also given an explicit opportunity to articulate shortcomings of the generated essays, noticing flaws in references and scholarly practice, an activity similar to peer reviewing work. Students dissected the generated essays and reflected on them using guiding questions, which helped them to discern and develop their own reasoning and arguments.

Were there other impacts on Phil’s understanding or practice?

In comparing how straight-forward it was to integrate interaction with GenAI on this module versus others that he leads, it has become clear to Phil that the amount and type of student engagement with GenAI is going to need to vary between modules, and certainly by subject area.

In creating these assessments Phil also came to think about how to adapt activities if any students did not want to use GenAI tools themselves, for ethical reasons. For the essay assessment he provided some pre-generated essays that could be used so that there was the ability to opt-out of using the tools while still engaging intellectually with the products of GenAI.

How did conversations go between Phil and his students about how GenAI might impact their learning?

Phil has run this module and these assessment for 2 years so far, and he has noticed a significant change between these two years in students’ prior experience of GenAI. In the first year when talking to the students they looked nonplussed at the mention of GenAI tools; no one said that they had used them during A-levels, and no one said they had friends who had used them. The second year, it was completely different. The students not only knew about GenAI tools but said they had already started to “dabble.”

Has Phil made changes to his assessment, or does he plan to?

Phil has observed that development of this technology seems to be changing very quickly, and so while he does intend to review and revise his assessments, he is going to wait until much closer to the module start to do so.

One area where he’ll be looking to make improvement is in the assessment instructions. As the types of assessment activities his instructions describe are not very familiar to his students, he has been providing clarification during lectures and seminars, but as this doesn’t reach those who didn’t attend in-person, he will need to integrate these clarifications into the assessment briefs themselves.

Given the dramatic increases in prior knowledge of GenAI between the first and second years, in the coming year Phil is expecting to see students who are even more experienced with the use of GenAI tools before they begin his module. He has been guiding students through writing effective GenAI prompts but wonders if students’ prompting skills might have outstripped his own when he next runs the module.