November 11, 2024
With the emergence of new artificial intelligence (AI) applications and programs, the uses of AI have been a hot topic across societal sectors – healthcare, business, and notably, education. AI is framed as a means for transforming educational settings, such as in how administrators complete tasks, how instructors plan and teach their curriculum, and how students learn and are assessed. But what if the uses of AI in the classroom are not as straightforward as they seem?
Raquel Coelho, assistant professor in the Department of Informatics and Networked Systems and research scientist at the Learning Research and Development Center, and Aakash Gautam, assistant professor in the Department of Computer Science and Department of Information Culture and Data Stewardship, are cautiously optimistic about AI in classrooms and education more generally.
Responding to the increased hype about AI, Coelho and Gautam each urge caution against viewing AI as a breakthrough educational solution, and instead emphasize the importance of human-centered design of AI. Although AI can help delegate and automate task completion, the role of human educators is essential for fostering conditions necessary for a productive and meaningful learning environment.
“If we look at history, we see that there has always been a lot of enthusiasm about education and technology, even with claims that technology will completely transform education,” said Gautam. “But we have also learned that the human side of education is more important than anything else. It is this human side that we need to consider deeply.”
Mainstream controversies about AI in the classroom typically offer two stances: either completely for AI, or completely against AI. Coelho and Gautam argue that instead of taking sides, we should focus on how AI is being designed and for what ends. Technology is designed to establish order in the world, but ideas of what order means can differ. When AI is being used to educate a new generation of thinkers, it is essential that AI is designed responsibly, with people’s wants, needs, and backgrounds in mind, and with critical attention to ethical issues of justice, fairness, bias, and privacy.
“We are dealing with young people who are still developing,” said Coelho. “Learning about and serving students’ needs must be front and center. Their concerns should motivate responsible and ethical design.”
In addition to paying careful attention to AI design, Gautam suggests that the time and resources invested into AI can be distributed elsewhere to contribute to more productive, collaborative classrooms.
“If students begin to rely on AI to learn, they are going to think of classrooms as a place where they have to perform, like for a grade,” explained Gautam. “It takes away the value of education - which is to support people to know themselves and understand their surroundings so they can have the power to shape it. We need to invest that time and money that’s going into AI into human infrastructures that enable such a space, like teacher professional development and better pay for teachers.”
As the popularity of AI continues to increase, the time to make responsible decisions about AI is now. AI holds great promise to be useful in education, if its applications are designed in a human-centered way and used to augment human cognition rather than replacing it.
“Early design decisions are difficult to change later on, as they become the foundation for all the infrastructures built around them. This is why we need to spend considerable time thinking about these initial choices,” said Coelho. “Many critical AI design decisions are being made right now: socio-technical infrastructures, policies, and massive environmental demands of AI factories. I insist that we prioritize community driven critical design and assessment of risks of AI before it goes any further.”