CITRIS experts weigh in on the ways the spring’s hottest machine-learning interface has the potential to transform education.

Snapshot

ChatGPT, a machine learning interface, has been widely covered since its debut with many claiming it will change education.

Brian Christian, a visiting scholar at UC Berkeley, believes it will be "exhilarating, destabilising and transformational." The technology will upend course assignments, become a 24/7 tutor and change the way knowledge work is done in higher education. To prepare students for the changing job market, educators need to adapt to the changing requirements of employment in society's automation of knowledge work.

Brandie Nonnecke, founding director of the CITRIS Policy Lab and associate research professor at Berkeley, is helping lead the effort to ensure equity and accountability are prioritised and risks minimised for the technology. She said the question now is how to set appropriate guardrails to ensure students can learn from these types of models. Berkeley scholars argue that while ChatGPT has promise, plagiarism detection programs like Turnitin have already said they will incorporate programs to detect AI-generated, ripped-off text.

It is more complicated than that…

Others have criticised ChatGPT as the death knell for learning. Berkeley scholars say it is more complicated than that. Nonnecke said burying our heads in the sand and saying “you can’t use them at all” isn't the answer. Rather, students need to learn how to meaningfully use these tools.