JAPAN Universities take lead in forging AI policies, practices

suvendrini kakuchi   

Japanese universities, latecomers in the use of digital technology for teaching, are fast establishing policies on the adoption of artificial intelligence in the classroom. Experts predict that utilisation of the technology will steer important reforms in higher education, particularly around redesigning current teaching methods that lack discussion and focus on lecture-style instruction, inhibiting individual learning.

“While the adoption of generative AI is still at the early stages in universities, the initial results indicate its transformative role in teaching styles that encourage more debate on its use,” said Professor Miyashiro Esteban, who teaches business management at Aichi Toho University, a private institution located in Nagoya, Japan’s third largest city.

In addition, he explained to  University World News , professors are beginning to view AI as crucial to reduce administration workloads that are a burden on their time. “Freeing up much needed time for their own research has indicated the plus side of AI.”

In his classes Esteban uses bots such as ChatGPT, Google Bard and Microsoft HHS Strategy and Solutions. “My teaching encourages my technology-savvy students to understand the benefits of using these tools for their learning while having discussions on the ethical perspectives of AI,” he said.

Experts say that there is widespread concern in Japan about AI ushering in an era of rampant cheating in education, such as using ChatGPT to write essays.

Nevertheless, more universities are embracing the technology as the way forward, as does Esteban: “AI is a fantastic tool when used for the purpose of opening up learning and research and students need to be taught that,” he said.

AI guidelines for universities

On 13 July 2023, the Ministry of Education released a document with guidelines for universities to develop their own AI policies.

Overall, the ministry pointed to the importance of responding to AI proactively, for instance by encouraging independent learning for students in terms of information gathering and programming assistance. The document also permitted AI use by academics to develop teaching materials, and in university administration.

On tackling misuse, the ministry said it was “inappropriate” for students to “entirely copy” AI text output for their assignments.

In order to further strengthen monitoring, the ministry requested lecturers to ask students to specify the type and content of the AI used. In addition, instructors are required to conduct oral examinations as part of the students’ work evaluation.

Given the experience of having to close facilities for almost two years from 2020 during the COVID-19 pandemic, when Japan scrambled to expand online learning, universities are currently taking steps to embrace AI by adopting their own policies.

Preventing student reliance on AI

The main thrust is expanding the use of AI but with initiatives to curtail a disruptive digital era. Against this backdrop, a clear goal is to prevent students relying too heavily on technology.

According to a June survey by Asahi newspaper, nearly 33% of college students in Japan used ChatGPT, with 14% reporting that they have used it for assignments. The same results also indicated that students are aware of the potential risks of the chatbot.

Regarding the way they use ChatGPT, almost 78% of students replied that AI was important for improving their writing and 70% said it improved their thinking. A high number – 91% – said they checked the answers given by AI and also added their own ideas to their assignments.

Universities ahead of the game

Nagoya University, a leading national higher education institute, enacted its AI policy last year. It states the importance of using generative tools for learning, including ChatGPT, the text and image app that is most popular with students.

In March this year the university’s president, Naoshi Sugiyama, gave two congratulatory speeches to graduates – one his own and the other created by ChatGPT.

After reading the generative AI speech, Sugiyama pointed to the expanding use of ChatGPT for language, images and voice, and urged students to “keep asking yourselves what your own values are and challenge yourselves”.

Professor Yasuo Kuniyoshi, director of the Next Generation Artificial Intelligence Research Center at the University of Tokyo, explained in an interview that prohibiting ChatGPT was not the popular view in Japanese universities. “The more effective way to respond to the age of big data is teaching students to enhance their capabilities by exploring the many possibilities of the power of AI under close monitoring,” he argued.

Kuniyoshi said his centre goes beyond existing frameworks for AI to create new technology systems in the field of intelligence science that are truly helpful to humanity and could become a driving force of society in the future.

“The recent AI boom becomes a trigger to rethink society in a world that is unpredictable with the growth of new technologies. It is necessary, therefore, to work on projects that employ the cooperation of a multifaceted research approach to find answers in this dynamic real world we inhabit,” he explained.

Kuniyoshi and his colleagues conduct research using bots with graduate students. They explore answers to questions such as what kind of society humanity should aim for and what new technologies and social systems will become necessary to achieve this.

In the Japanese higher education environment that is rooted in structure, Kuniyoshi believes the power of AI will spell a new era for university teaching and learning.

Bots will upend traditional emphasis on exams or lecture learning by flipping classrooms and encouraging students to find creative ways to collaborate with new technology. “This requires students to draw from personal experiences, reverting to oral presentations, and collaborating with diverse disciplines,” he said.