Pages

29 January 2023

ChatGPT Is Coming for Classrooms. Don't Panic


WHEN HIGH SCHOOL English teacher Kelly Gibson first encountered ChatGPT in December, the existential anxiety kicked in fast. While the internet delighted in the chatbot’s superficially sophisticated answers to users’ prompts, many educators were less amused. If anyone could ask ChatGPT to “write 300 words on what the green light symbolizes in The Great Gatsby,” what would stop students from feeding their homework to the bot? Speculation swirled about a new era of rampant cheating and even a death knell for essays, or education itself. “I thought, ‘Oh my god, this is literally what I teach,’” Gibson says.

But amid the panic, some enterprising teachers see ChatGPT as an opportunity to redesign what learning looks like—and what they invent could shape the future of the classroom. Gibson is one of them. After her initial alarm subsided, she spent her winter vacation tinkering with ChatGPT and figuring out ways to incorporate it into her lessons. She might ask kids to generate text using the bot and then edit it themselves to find the chatbot’s errors or improve upon its writing style. Gibson, who has been teaching for 25 years, likened it to more familiar tech tools that enhance, not replace, learning and critical thinking. “I don’t know how to do it well yet, but I want AI chatbots to become like calculators for writing,” she says.

Gibson’s view of ChatGPT as a teaching tool, not the perfect cheat, brings up a crucial point: ChatGPT is not intelligent in the way people are, despite its ability to spew humanlike text. It is a statistical machine that can sometimes regurgitate or create falsehoods and often needs guidance and further edits to get things right.

Despite those limitations, Gibson also believes she has a responsibility to bring ChatGPT into the classroom. She teaches in a predominantly white, rural, low-income area of Oregon. If just the students who have ready access to internet-connected devices at home can gain experience with the bot, it could widen the digital divide and further disadvantage students who don’t have access. So Gibson figured she was in a position to turn ChatGPT into, to use educator-speak, a teachable moment for all of her students.

Other educators who reject the notion of an educational apocalypse suggest that ChatGPT might not be breaking education at all, but bringing attention to how the system is already broken. “Another way of thinking about this is not how do you find new forms of assessment, but what are our priorities in further education at the moment? And perhaps they’re a little bit broken,” says Alex Taylor, who researches and teaches human-computer interaction at City, University of London.

Taylor says the bot has prompted discussions with colleagues about the future of testing and assessment. If a series of factual questions on a test can be answered by a chatbot, was the test a worthwhile measure of learning anyway? In Taylor’s view, the kind of rote questions that could be answered by a chatbot don’t prompt the kind of learning that would make his students better thinkers. “I think sometimes we’ve got it back to front,” he says. “We’re just like, ‘How can we test the hell out of people to meet some level of performance or some metric?’ Whereas, actually, education should be about a much more expansive idea.”

Olya Kudina has used ChatGPT as a tool in her own classroom at Delft University of Technology in the Netherlands, where she teaches graduate and undergraduate courses on AI and ethics. In December she gave her undergrads a debate-style assignment using ChatGPT. Groups of students first presented three arguments and two counterarguments, supported with academic references, to the class without AI assistance. Next they fed the same assignment to their choice of either ChatGPT or its predecessor GPT-3, then compared the chatbot’s answer with their own organically made text.

The students were dazzled by how quickly the chatbot rendered information into fluid prose—until they read it with a closer eye. The chatbot was fudging facts. When students asked it to back up an argument with citations from scholarly texts, it misattributed work to the wrong authors. And its arguments could be circular and illogical. Kudina’s students concluded that, contrary to fears of a cheating epidemic, copying from ChatGPT wouldn’t actually net them a good grade.

Kudina says that teachers should neither ban ChatGPT nor embrace the technology without question. She advocates for her profession to “critically appropriate” the technology and find more creative ways to collaborate with it. For example, students might use the chatbot to spark new ideas or arguments. (One of her students likened ChatGPT to a superpowered Google search.) Kudina thinks ChatGPT might also spur educators to get more creative with assignments, for example by designing them to draw from students’ personal experiences, information that ChatGPT couldn’t have picked up from its training data.

That’s not to say ChatGPT won’t be at all disruptive to education. The bot emerged at a time when many teachers are experiencing burnout after emergency remote learning during the pandemic. Now another technological phenomenon threatens to upend their entire approach to teaching, creating more work. And the student privacy implications of ChatGPT, particularly at the K–12 level, are unclear. OpenAI does collect some data on users and says it reviews conversations with ChatGPT; the company’s terms of service state that users must be 18 or older, although the bot doesn’t attempt to verify age.

Completely barring ChatGPT from classrooms, tempting as that may be, could introduce a host of new problems. Torrey Trust at the University of Massachusetts Amherst studies how teachers use technology to reshape learning. She points out that reverting to analog forms of assessment, like oral exams, can put students with disabilities at a disadvantage. And outright bans on AI tools could cement a culture of distrust. “It’s going to be harder for students to learn in an environment where a teacher is trying to catch them cheating,” says Trust. “It shifts the focus from learning to just trying to get a good grade.”

In January, at the start of the new semester, the New York City public schools banned ChatGPT on school devices and networks due to “concerns about negative impacts on student learning and concerns regarding the safety and accuracy of content,” a spokesperson told Chalkbeat. Marilyn Ramirez, who teaches high school English in Washington Heights in New York, says that her conversation with WIRED was the first she had heard of the ChatGPT ban in her district and that she was not directly informed by the New York City Department of Education.

Ramirez is the kind of teacher who will do a dramatic reading to get her kids, many of whom are special education and English language learners, hyped up about a Queen Elizabeth I speech. She’s not worried about ChatGPT. She makes an analogy with how she allows her English language learner students to use Google Translate but also helps them see where the technology falls short, and when it’s appropriate to use. She sees ChatGPT similarly: beneficial with a teacher’s guidance but ultimately limited.

No comments:

Post a Comment