Balancing AI With Our Values

Opportunity for creativity or a convenient shortcut? Gary Henderson explores how educators can ensure students are still learning and not just ticking off check boxes

The growth in artificial intelligence (AI) use in schools has sparked a heated debate about its implications on academic integrity and the nature of student learning. Many have focused on how we prevent or detect AI use; however, I suspect neither of these are reliably possible. So, if we accept that AI is going to be used by students, be this for coursework, homework, revision or other activities, then what are we to do?

Creativity and Convenience

We want students to be creative, including in how they demonstrate their knowledge and their learning. Therefore, something that may enhance creativity should be a good thing, and AI has the capability to enhance creativity. The fact I can quickly create images as prompts for a lesson topic, or even videos or virtual environments, or new programming code, all with such speed and convenience demonstrates how AI has made me more creative.

The ease and speed of AI allows me to explore and to iterate in a way that previously wasn’t possible, refining and developing ideas. I have shifted my use of time away from focusing on creation, such as drawing or developing code, towards ideation, as the AI can take on the creation element. But things aren’t quite that simple. Generative AI is a huge mathematical model and therefore tends to provide what is statistically likely based on the prompt it is provided and the training data it has ingested. Doesn’t that therefore make what it provides more generic and average and therefore less creative?

Is it Truly Creating?

I would suggest the answer to this question depends on the person using the AI and their intentions. If a student is using AI purely to get work done with the least amount of effort, then they will likely put little effort in and will rely purely on the output from the AI, which likely will tend towards being generic. They might simply copy and paste the task, as provided to them by their teacher, into their chosen AI solution before making some cursory word or phrase changes. A piece of work will be submitted but the student will have expended little effort and will have learned little.

If, however, the student is seeking to produce the best work they can then they are likely to refine and develop their prompts, providing additional information and resources to the AI, adding their views and perception and in doing so, influencing and developing the resultant output away from the generic. Through the use of AI, the students content creation will be easier as the AI can create a structure or individual paragraphs, providing revision suggestions and more. But this will allow the student to spend more time on ideation, on thinking about the topic and on being creative. I would suggest the student will have experienced desirable difficulty in the task as they explore and iterate, even although content creation wasn’t fully in their hands. I would also suggest that learning would therefore have occurred, albeit it may be slightly different learning than would have occurred without the use of an AI tool.

Learning and Skills Building

At the end of the day what we are looking to achieve is for students to have learned, and for students to build the skills to help them continue learning beyond formal schooling. We want life-long learners. We also want them to be able to do what they need to do, to the best of their ability. On both of these counts we need to recognise the implications of AI. In terms of the future, AI is only going to grow in its use, so we need students to be experiencing AI now. Plus, in terms of doing their best, why wouldn’t you use the tools available including AI? But this all boils back to the person, the student and whether they themselves are focused on learning, on doing their best, or whether they simply want to move on to the next task, the next stage, with the minimum effort, and correspondingly the minimum learning.

What Happens Now?

I think the key is for schools to establish how we work in partnership with AI, knowing that students are going to use it. If we operate from this proposition, that student use of AI is inevitable, then maybe we start to ask students about how they used AI in their work, what worked well and what did not. I think it is also important that we reframe how we speak about what students are doing; They aren’t doing coursework or doing homework, for the sake of coursework or homework. They are doing these tasks to demonstrate their learning. If therefore a teacher, knowing their class, is concerned that the presented work isn’t representative of the students learning then it is simply for the teacher to seek to validate the learning through another task or simply ask the student questions as to their submitted work. Maybe the solution to AI is more human interaction with our students?