Students are “creatively” using new technologies in ways their lecturers do not understand but know that it is “not good for them” to use artificial intelligence to simply do their assessments for them, according to the head of education at Google.

Shantanu Sinha, Google’s vice-president for education, told 51�Թ� that AI is “transforming everything”, but conversations around student cheating may be misguided.

In recent months Google has released a series of products targeted at students and educators. Its Notebook LM can transform students’ notes into flashcards, quizzes, mind maps and even podcasts and videos with voiceovers and animations. Earlier this summer, it also introduced its “guided learning” mode on Gemini – similar to Chat GPT’s Study mode – which acts as an electronic personal tutor to guide students through their studies, rather than providing them with the answers. 

Sinha said that this guided learning mode was actually something pushed by students, who acknowledge, “I know it isn’t good for me when I just ask a chatbot and get the answer and copy that down”.

51�Թ�

“They know that’s not good for them because they know that’s jeopardising what they’re going to get assessed on. Ultimately, their motivation isn’t just that assignment. Their motivation is getting a good grade and doing well.”

Sinha believes that students are using these technologies in creative ways beyond the imagination of lecturers, and the tools have the capacity to create more “self-motivated” learners by making learning more engaging.

51�Թ�

“The pessimistic way to look at this would be like, ‘OK, students are cheating now’. But it’s not clear that’s actually true.”

Instead, the rise of AI potentially highlights areas where universities can improve the effectiveness of their assessments and ensure that they are driving critical thinking.

“For any professor, you have to look through what you’re [assessing] in your assignments to try to understand where there’s a risk that students might be bypassing them inappropriately, and then question, ‘how do I actually drive students to the places where you’re getting learning that really matters?’”

He added, “If your assignment is such that you could look things up that easily, then maybe that’s not the best assignment. There’s more to learning than that.”

51�Թ�

As universities evaluate their AI policies, he said it was “pretty hard” to have “a one size fits all approach” on what’s acceptable, noting that there are different approaches and opinions on AI among academics. 

However, he said “more clarity” would be useful for students, who “don't want to use it the wrong way, but sometimes they question, ‘why was asking my friend for help OK’ or ‘why was searching on the web OK’ but this is not. Those lines are kind of blurry”.

Although AI is having a profound impact on higher education, the “core of the classroom” still revolves around discussions between students and academics, as well as written exams, “which are totally AI-proof”.

For Sinha, it’s a positive future – and one where higher education can be at the forefront of AI innovation and creative approaches to technology and learning.

51�Թ�

“In more and more places, it is important for higher education to be at that cutting edge, because AI is going to transform almost all fields to an extent.”

He continued: “We can make learning even richer than it was before, and we can start to create project-based learning that leverages AI to unlock all kinds of possibilities, and you can start to push much deeper and creative experiences in classes. It’s going to take some time, and there’s a lot of innovation that still needs to happen to get there but I think that there’s huge potential.”

51�Թ�

juliette.rowsell@timeshighereducation.com