AI tools are no longer niche technologies, but are being widely used, discussed and publicised. The University Group has acknowledged that; ‘we need to be future-focused and ensure we take advantage of what AI has to offer […] However, whilst taking advantage of what AI has to offer our students, we need to ensure we educate and guide students towards appropriate use of AI.’ [Artificial Intelligence – staff guidance]
It is therefore important understand the ways that the technology could potentially impact teaching and learning. Below are a number of areas that require consideration and, in line with the University’s stance, take a balanced approach of identifying opportunities while acknowledging areas of challenge and/or risk that may need mitigating against.
Embedding AI into curriculum
AI is already affecting many areas of wider society, including transformational change within certain industries and sectors. By integrating the use of AI into teaching and learning (rather than simply teaching ‘about’ AI) we are stretching and challenging students to adjust to the future where AI assistance is normal.
Unesco, in their quick start guide, ChatGPT and artificial intelligence in higher education (Unesco, 2023), provide a summary table of potential implementations, this is included below:
(Unesco, 2023)
Greater focus on graduate outcomes
Integrated use of AI tools can help improve students’ skills beyond subject knowledge. Learning about AI’s impact can enhance their abilities and positive behaviour towards emerging technology. Using them selectively could help in developing graduates’ attributes.
Streamlining administrative processes
ChatGPT can also help streamline many time-intensive administrative processes, such as conversion of notes into prose, auto-generation of meeting minutes from closed-captions etc. Less time on administrative tasks could result in more time on more valuable activity. These benefits are not just limited to academics but affect all University staff.
Personalised learning
There is scope for AI to greater increase the personalisation of the learner journey. Teaching materials can be generated and tailored to the individual’s needs, interests and experiences in a way that would not be feasible if curriculum is only produced and prepared by a relatively small number of academics.
Plagiarism/submission of non-original content
The obvious threat of using AI within the HE sector relates to students using it to generate non-original content. However, this can be mitigated through careful consideration of assessment methods. We have addressed this in more detail here [link to article 3]. AEPD have also produced resources about assessment design to counter academic misconduct and can be accessed here.
There are some tools already available that have been designed to detect AI generated content. However, our internal testing suggests that these are not yet reliable for drawing definite conclusions around AI use.
Increasing Digital Divide
As AI gains a foothold into the everyday lives of staff and students, it will become even more important to ensure that all have access to these tools. Digital poverty and the digital divide have been a genuine concern for a long time, but lack of access to essential AI tools could result in even greater disadvantage for those without easy access to these technologies.
Skills-loss
Although AI usage has the potential to build up new skillsets, there is a risk that other skills will be lost in the process. A bit like ‘sat-nav syndrome’, or reports that many young people are unable to read analogue clocks, will students lose the ability to write without AI assistance? One of the most common use cases of AI is to generate a starting point for written work, the focus then becoming the process students engage in to refine outputs and ensure the final outcome is accurate, balanced and appropriate. However, there is value in the resilience and creative thinking required to start with a blank page and create something original, and this kind of thing could be under threat with the advent of ChatGPT and similar tools.
Encourages focus on outcomes-driven learning
Many of the current use-cases (and benefits) of ChatGPT usage are around streamlining of processes and saving time. These learning ‘short-cuts’ have the potential to encourage outcome-driven learning, where students are predominantly focused the achievement of qualification over the process of learning. Careful consideration will have to be given to how AI can better engage learners to develop an increased interest in their chosen subject, rather than just providing a path of least resistance and to also encourage a greater focus in learning design on skills and capabilities rather than knowledge. (See personalised learning opportunity above as an example).
Important considerations when using or encouraging the use of AI tools
As previously stated, the University Group’s position is to ensure we are taking advantage of what AI tools have to offer. However, their application in teaching and learning environments is not without important considerations.
Firstly, it is critical to consider the contexts in which it is safe to use AI assistance. Below is a Flowchart devised by Aleksandr Tiulkanov, AI and Data Policy Lawyer, (Unesco, 2023). It was designed as a quick guide to determine if it is safe to use ChatGPT. However, this can be applied to many generative AI tools:
(Unesco, 2023)
Finally, any context in which AI is being used within teaching and learning should be aligned to the University’s position on such tools. The staff guidance for AI outlines appropriate and inappropriate uses of these technologies.
References
OECD (2023) ChatGPT and artificial intelligence in higher education: a quick start guide, OECD https://unesdoc.unesco.org/ark:/48223/pf0000385146
Article Authors
David Ashworth and Martin Jenkins