The launch of ChatGPT and other large language models (LLMs) sparked a wave of panic in the academic world, with some lecturers going so far as to claim that the end of the traditional college essay was nigh. Others were more sanguine, pointing to the technology’s capacity for hallucination as a source of comfort and contending that ways would be found to identify machine-produced essays.
Regardless of which viewpoint you subscribe to, the fact remains that the generative artificial intelligence (GenAi) genie is now out of the bottle, and it is not about to go away. The issue for academics and others is how to accommodate this new fact of life.
UCD School of Business associate professor Alessia Paccagnini believes academics need to embrace the technology and incorporate it into curriculums but is also keenly aware of the challenges it presents.
“Generative AI is a big threat to academic integrity because it makes it so easy for students to write essays, reports, and other homework without putting in any real effort and their personal contribution,” she explains. “This leads to plagiarism because most tools that look for plagiarism need help finding content that was made by AI. GenAi assignments also keep students from learning important skills like research, writing and thinking critically because they skip the learning process. This false representation of ability not only makes academic work less reliable, but it also gives students who use AI an unfair edge over those who don’t.”
But GenAi could be useful if it is used in a smart way, she adds. “It can be used as an extra resource to help students understand difficult ideas and get better at writing and researching. To lower the risk of AI being misused, the education system can change by creating new ways to test students that focus on critical thought and active participation in class. Adding AI literacy to the curriculum can also teach students how to use AI in an ethical manner. GenAi can be used as a useful tool for learning instead of a danger if these steps are taken. For example, students can use AI to summarise some facts and concepts.”
Teachers and professors should start to include AI in their module description, helping students to exploit this useful resource to improve their learning experience, she contends. But they also need to set boundaries to ensure that AI does not worsen the learning experience.
Alan Smeaton is professor of Computing at DCU and is a member of the Government’s AI Advisory Council. He agrees that the technology needs to be embraced but accepts that some may struggle with this.
“When it first came out there was panic,” he says. “Some countries wanted to ban it. Then they tried to come up with ways of detecting GenAi creations. That’s not going to work. The GenAi that appeared one and a half years ago was quite primitive in comparison to what we are using today. It is an order of magnitude more powerful. The hallucination problem is basically solved. Universities have been able to react and pivot. Lecturers can design GenAi into assessments.”
He also makes the point that GenAi will be available to the students in their working lives so it should not be ruled out when they are in college. “You can design it out by doing assessments in a controlled environment,” he says. “But is designing it out a fair way to educate students? Is designing it in a better way to do it? Designing it in is absolutely the correct way. The pace of change in GenAi means that education is going to have to move at a pace it’s not comfortable with. The third-level system is more nimble and agile. It will be much more difficult at second level. We know how difficult it is to change the second-level curriculum.”
GenAi is part of a new educational reality, and educators must respond, Paccagnini adds. “We are living in a constantly changing world, and students should be exposed to innovative teaching methods and modern assessment tools. The traditional teaching style is no longer efficient; AI, new technologies, computers and social media are becoming as integral to learning as traditional books. Therefore, new ideas for assignments, which are more inclusive and effective for future work environments, are needed. Writing essays is no longer the most efficient way to assess students.”