Australian universities declare war on ChatGPT

Major Australian universities have altered the way exams and other assessments are conducted amid concerns that students are using ChatGPT to write essays. The Guardian reports.

the Group of Eight revised the assessment process amid new technologies. Under the rules, the use of AI is treated as fraud.

According to Matthew Brown, the group7s deputy executive director, institutions are “actively fighting” AI-based solutions.

“Our universities revised the assessment process in 2023, including proctored exams, […] broader use of written and paper-based assessments, […] and tests only for units with a low risk of misconduct,” he said.

Brown added that the shift in assessment approach is “crucial,” as institutions seek to “get ahead of developments in AI.”

In the University of Sydney’s updated academic integrity policy, it explicitly states that “creating content using artificial intelligence” is fraud.

According to a university spokesperson, some students have already used the technology to write assignments.

“We also know that AI helps to acquire knowledge and becomes part of the tools we will use in the future. Therefore we need to teach our students how to use it legally,” he said.

The Australian National University has changed its approach to assessment. The institution now relies on laboratory and fieldwork, and introduces more oral presentations.

Other Group of Eight institutions have also expressed concern about the rise of text generators, especially ChatGPT.

Toby Walsh, an AI professor at the University of New South Wales, said instructors have been involved in developing approaches to counter text generators. He said students are already using them to write essays.

“We should have anticipated this. … We are moving into the future with little awareness. But this is a step-by-step shift — ChatGPT is available, has a user-friendly interface, and it is easy to experiment with,” said Walsh.

The professor added that with more advanced programs, including OpenAI’s GPT-4, simply banning the platform is unrealistic.

To counter counterfeits, Walsh proposed using technical measures such as watermarking. However, he noted that they can be removed with other software.

“This is an arms race, … and you will never win,” said Walsh.

The researcher also believes that AI holds great potential for innovation and optimization in education. He suggested that such technologies could help instructors assess essays and generate feedback on them.

“We do not want to destroy literacy, but have calculators destroyed the ability to count?” said the professor.

In January 2023, public schools in New York banned the use of ChatGPT on school devices and on internal Wi-Fi networks.

In October 2022, Reddit shared stories of how they used language models to complete homework.

Subscribe to ForkLog AI news on Telegram: ForkLog AI — all the news from the world of AI!