How To Tackle The Challenges of Generative AI in Education
Generative AI solutions such as ChatGPT and AutoGPT have been embraced by various sectors due to their ability to quickly come up with precise answers and summaries.
However, the use of these tools has sparked concern amongst academic heads in higher educational institutions. The worry stems from the potential negative impact on pedagogy and learning if students use these tools to generate content for their projects. This defeats the purpose of giving assignments that are meant to develop research and original thinking habits.
To tackle the issue, institutions such as Azim Premji University are exploring ways to deal with generative AI tools like ChatGPT effectively. This includes implementing measures to prevent its inappropriate use, such as plagiarism. The university has also started an awareness campaign amongst students and members of the university and is experimenting with different types of assessments to limit the influence of these tools on the learning process.
However, some believe that generative AI is not yet a significant threat to the IIT ecosystem, as its curricula is not solely based on text book-based evaluation. Instead, the pedagogy and curricula are designed to develop intuitive thinking, creativity and innovative problem solving skills.
There is much debate in academic circles worldwide on the topic of generative AI solutions. The challenge posed by AI and automation has been discussed at an international meeting of Vice-Chancellors, with the consensus being that pedagogies will be impacted significantly by generative AI solutions. KL University has decided to focus on activity-based, project-based learning and gamification, which previously made up just 10 per cent of the learning method, but now could go as high as 60 per cent.
Rajiv Tandon, CEO of Bits Pilani WILP, warns that knee-jerk reactions should be avoided when tackling the challenge of generative AI, and that progressive institutions should first deepen their understanding of AI's opportunities and implications on teaching, learning, and assessments.
Finally, Tandon suggests that educators should create "wicked problems" for projects. These are problems that are not easily solved by just knowing past patterns, which AI often uses to build solutions. Solving these problems requires the learner to demonstrate an understanding of the interconnections between phenomena to design their approach to solving them. This more difficult approach can prevent AI tools from being a simple answer to academic assignments.
