This page has been created to house the latest university-supported resources and information related to the impact of Artificial Intelligence on learning and teaching. It will be updated regularly to reflect relevant and timely information from across UOW and the sector to capture the evolving conversation and practices on the topic.
"AI will increasingly become endemic in our work and everyday lives. We shouldn’t be scared of it. Instead, we must try to understand and embrace it while constantly re-evaluating the moral and pragmatic implications of this revolutionary technology. "
- Theo Farrell, Deputy Vice-Chancellor (Academic & Student Life)
To start the conversation, Deputy Vice-Chancellor (Academic & Student Life), Professor Theo Farrell, shares his thoughts below.
Theo Farrell: Hi, colleagues. I just thought I'd offer a few thoughts about ChatGPT and the new suite of generative AI tools that we're seeing emerging now. Obviously there's concern in some quarters about what are the implications for exams and assessment at UOW and indeed universities across the world. So I'm speaking to you, it's early 2023. We've had ChatGPT on the scene now for a couple of months. Look first thing to say straight off the bat is, as many of you'll know nothing new in AI, AI is endemic already in our world, it's all over the place. If you use Siri or you use Alexa at home, that's AI-generated content. If you use Google search, that's AI-generated content. But of course, ChatGPT in many respects is new and different because, you know, if those of you who may have used it you put, give it a question, it gives you an essay-length response.
And so obviously as academics, the first question we perhaps ask ourselves is, oh, goodness me could a student use this to produce work? So in the university, we are doing a whole bunch of work around how we can better assure our assessments and exams and so forth going forward. And it's, you know, it's gonna be a bit of a learning journey for us because obviously as the tools evolve and they will evolve quite quickly we're gonna have to respond as we go along. And that's all important. And, and so we are working on that and we will work with you on that. But there's a larger play here, which I want to encourage everyone to keep in mind which is an exciting world that is unfolding in front of us. So you know, let's just take it back to what is our purpose at UOW. Because I want to remind all of us of our purpose and goal number one of our strategy is to empower our students for their future. Now ChatGPT and the tools that are like it are giving us a glimpse, perhaps more than a glimpse it's opening a window to the future.
And that is a future that is profoundly shaped by AI. It's going to reshape our economies with all sorts of, obviously, implications for the jobs our students will be doing, and it's gonna reshape our societies. And obviously, as a university, we're driven to shaping a future that benefits as many people as possible. So as we think about the what we're teaching and how we learn and how we assess this year and in the coming years, what we want to do is harness all of the creativity, all of the opportunities that AI tools like ChatGPT are now making available. The current tools like ChatGPT, they're not perfect one of my colleagues said it's somewhere between poor and brilliant. When you use the tool, you'll see some, some of the stuff it produces, it's a bit mediocre, but there, some of the stuff is quite, is quite amazing. So what I want to do is encourage us as educators and as the university community to embrace this technology and to play with the technology and to integrate it in the way that we teach. It's immensely creative.
And of course, we have to have conversations with our students around the ethical use of the technology. That's really important part of preparing them for their future. So we shouldn't be afraid. We need to embrace this technology if we are to prepare our students for their futures, and obviously we want to do so in a responsible way that's going to shape a beneficial future for as many people as possible. It's a journey we're going to go on together with our students. I'm really looking forward to that journey. And I, I hope you do too. Thank you.
Click on the headings below to expand the sections.
What are Artificial Intelligence tools such as ChatGPT?
ChatGPT is only one of a range of Generative Artificial Intelligence (GenAI) technologies developed by OpenAI and being built on by other developers due to the provision of open access to the underlying code. Trained on large datasets, GPT stands for Generative Pre-trained Transformer to create language generation models, GPT tools can be used to generate text, images, music, code, data and videos in addition to evaluating input, organising data, and recommending strategies using an AI language model that can generate output similar to that of humans when given a prompt.
Impact of AI tools on assessment, learning outcomes and curriculum
One of the challenges faced by educators centres on the fact that AI tools may be able to provide passing level answers to some of the assessment tasks we are assigning our students. Consequently, the question arises of how we can assure that students have achieved the learning outcomes if they have used an artificial intelligence tool to generate their assessment submission.
ChatGPT and other AI tools developed via this technology are developing rapidly, so what we know on a day to day basis to be a limitation (such as its inability to confidently generate existing references at the time of writing) is likely to be addressed in a newer version.
When it comes to detecting AI-written materials, the challenges in detection are even bigger. At the time of writing, the OpenAI classifier, for example, cites 26% accuracy in detecting AI-written materials (OpenAI, 2023b). Turnitin (Turnitin, 2023), GPTZero, ZeroGPT and other tools are being actively developed to determine if the material has been AI-generated. Even though these tools will continue to grow in accuracy, they are unlikely to be a perfect detection mechanism and should not be considered solutions in detecting AI use in student work.
It is important to remember that using Artificial Intelligence in education does not automatically equate to misconduct and that there are ways AI can be used ethically for teaching, learning, and assessment (Eaton, 2022). At UOW, our focus institutionally is embedding the benefits of AI in curriculum while strengthening a culture of Academic Integrity to minimise the risks of Artificial Intelligence being inappropriately used in assessment to determine the achievements of learning outcomes.
"The future of academic integrity must rest not on outside technologies, but on the relationships between institutions, staff, faculty, and most importantly students."
- Rettinger & Bertram Gallant, 2022, p.95
Artificial Intelligence and opportunities for assessment
ChatGPT and its derivatives have the potential to revolutionise the way assessments are designed and administered. Assessments are alive and well and we now have this opportunity to make the most of emerging technologies to ensure that our interactions with students focus on the human element and on higher-order thinking skills. This approach will enable students to better navigate a world where artificial intelligence tools are going to revolutionise how we study and work.
In terms of assessment design, we need to make sure that the assessments we design and implement are quality assessment tasks (See UOW Assessment & Feedback Principles). This has not changed with the emergence of Artificial Intelligence tools; The difference is that we, as educators, can also use emerging artificial intelligence tools to assist us in designing assessment tasks for job-ready graduates. In short, “redesigning assessments to broaden the scope of allowable behavior acknowledges the technological developments of the 21st century and serves to reduce mis-conduct and increase learning" (Bertram Gallant, 2008, as cited in Rettinger & Bertram Gallant, 2022, p. 94).
These developments provide an opportunity to help our students develop the digital literacy needed in this fast-evolving space while maintaining relevance to changing organisations.
Experimenting with artificial intelligence tools like ChatGPT can generate results that are creative, funny and quite surprising. There may have also been moments when we were left unimpressed with an AI-generated response. The skill to conduct a critical and evaluative judgement of AI outputs is something that we want our students to develop to remain relevant as graduates of UOW. We want them to be able to decide how to use it legally, ethically and responsibly and we can assist our students to do that, by embracing technology.
Academic integrity is a core value at UOW and it is important for us as educators to have open and honest conversations with our students about the importance of acting with honesty, fairness, trust, and responsibility. These are values that are expected of the UOW academic community and are also reflected in the ethical code of conduct of most, if not all organisations and professions students will engage in after graduating. Research indicates that connecting academic integrity with the culture and “communicating that connection to students in clear, personal, and positive ways can dramatically curtail academic misconduct” (Rettinger & Bertram Gallant, 2022, p. 93).
The Vice-Chancellor has approved an amendment to the Academic Integrity Policy to provide the University with immediate coverage for the use of AI tools like ChatGPT. The definition of academic misconduct now includes “using work (e.g. assignment, essay, exam paper, research paper, creative project, data) generated by an artificial intelligence (AI) tool in an assessment unless expressly permitted to do so and with proper acknowledgement”.
AI tools are quickly developing, and a further review of the Academic Integrity Policy and Academic Misconduct Procedures is in progress – UOW staff will be invited to comment on any changes.
UOW guidance for students that can be used in your teaching is available via the Learning Co-Op resources on referencing and an FAQ on the topic of "Can I use ChatGPT and other AI?".
References
Eaton, S. (2022, December 9). Sarah’s thoughts: Artificial intelligence and academic integrity. Learning, Teaching, and Leadership. Retrieved February, 03, 2023 from https://drsaraheaton.wordpress.com/2022/12/09/sarahs-thoughts-artificial-intelligence-and-academic-integrity/
Farrell, T (2023, February 1). We can’t ban ChatGPT from education so we have to use it. The Australian. https://www.theaustralian.com.au/higher-education/we-cant-ban-chatgpt-from-education-so-we-have-to-use-it/news-story/40478992ac49541acedb36dd7fa33bdb
Fyfe, P. (2022). How to cheat on your final paper: Assigning AI for student writing. AI & Society, 1-11. https://doi.org/10.1007/s00146-022-01397-z
Lei, M., & Chen, Y. (2020). The limitations of AI detection software in academic integrity. International Journal of Advanced Computer Science and Applications, 11(2), 136-140. https://doi.org/10.14569/IJACSA.2020.0110221
OpenAI, (2023a). Build next-gen apps with OpenAI’s powerful models. Retrieved February, 06, 2023 from https://openai.com/api/
OpenAI, (2023b, January 31). New AI classifier for indicating AI-written text. Retrieved February, 03, 2023 from https://openai.com/blog/new-ai-classifier-for-indicating-ai-written-text/
OpenAI, (2019, February, 14). GPT stands for Generative Pre-trained Transformer, a transformer-based language generation model. Retrieved February, 03, 2023 from https://openai.com/blog/gpt-2/
Rettinger, D., & Bertram Gallant, T. (2022). Thirty years of active academic integrity research and engagement: What have we learned? Journal of College and Character, 23(1), 92-95. https://doi.org/10.1080/2194587X.2021.2017976
Turnitin (2023, January 13). Sneak preview of Turnitin’s AI writing and ChatGPT detection capability. Retrieved February 03, 2023 from https://www.turnitin.com/blog/sneak-preview-of-turnitins-ai-writing-and-chatgpt-detection-capability
