Rise of ChatGPT sparks conversation about educational values


Orion Kim

SECOND GLANCE. While this photo might look like a student staring at a computer, look closer and recognize that it is slightly distorted. The image was generated by OpenAI’s DALL-E 2 based on the prompt: “Create a realistic photo of a student using ChatGPT.” (Chat GPT photo)

Open AI released ChatGPT Nov. 30, 2022 an artificial intelligence chatbot that immediately took the world by storm. Within weeks, the world embraced this new technology: students used it to write essays, singers asked it to generate song lyrics, and home cooks used it to make new recipes.

If this magical chatbot can pull the answers to virtually any question out of thin air, what’s the danger? Academic Dean Tom Anderson believes that AI technology, while it can be helpful in some ways, poses a threat to students.

“If a student feels like they can find a shortcut that is less traceable, they might take that path,” he said. The danger is that taking shortcuts might damage a student’s ability to learn important life skills.

A student’s “capacity to gather, arrange, synthesize, and analyze information is going to be important in any field that [they] are in,” Anderson said.

In the long term, if students sacrifice opportunities to practice those skills they’ll eventually fall behind.

So what exactly is ChatGPT? Ask the program, and this is the response:
“ChatGPT is a large language model developed by OpenAI that has been trained on a vast corpus of text data to understand natural language and generate responses to a wide range of questions and prompts. You can think of it as a virtual assistant that can provide information, answer questions, or have conversations on a variety of topics.”
Although ChatGPT might seem flawless, it is far from completely polished. The previous paragraph seems clunky and robotic, which makes it distinguishable from human writing.

According to Anderson, AI technology still has a lot of room for improvement.

“I don’t think that the current ChatGPT is particularly strong. It has a very formulaic approach. I’ve already seen examples that kids have turned in that were clearly created by something like it,” he said.

Although there are limitations, Chatbots aren’t going to stop progressing. Many other search engines have already followed suit, including Microsoft’s Bing and Google’s Bard. According to the Harvard Business Review, chatbots will only continue to develop “AI has the potential to take over certain roles traditionally held by humans. As [it] continues to improve, more and more current jobs will be threatened by automation.”

Even though ChatGPT might seem like the next matrix, it isn’t exactly detrimental. According to the review, “The question isn’t whether AI will be good enough to take on more cognitive tasks, but rather how we’ll adapt.”

Senior Dylan Tan Stephenson and his neuroscience project group adapted strategically. Instead of using the technology to cheat, they decided to make a point.

The group completed the majority of their project without the help of AI technology. However, when the time came to assign a homework project for the class, they used ChatGPT to demonstrate its power. “It was actually amazing what ChatGPT could do. It came up with fake names, fake dates, fake sources, [cited] APA, which was amazing because you’d think it would only be able to write primitively,” Stephenson said.

After the presentation, they revealed that the project was AI-generated, and the class was shocked by what the technology was able to create. “People thought it was cool, but I don’t think they thought about what AI can do if people want to use it for malicious things,” he said.

Junior Violet Benson also found ways interesting ways to use the technology. “It can be useful to use [ChatGPT] to get basic background information and an overview of the topic,” she said. Additionally, there’s an abundance of creative ways to make use of the technology. “I know someone who used it to try and generate a new cooking recipe. You can also generate art ideas or just play around with it,” she continued.

It was actually amazing what ChatGPT could do. It came up with fake names, fake dates, fake sources, [cited] APA… you’d think it would only be able to write primitavely.

— Dylan Tan Stephenson

Despite the potential benefits of ChatGPT, its proficiency makes it dangerous. For many students, it’s difficult not to use chatbots as the easy way out of an assignment.

According to a survey from study.com, 89% of students have used it to complete a homework assignment.

According to the study: “While ChatGPT can be useful for generating things like writing prompts or assisting in outlines for students, the students themselves may be more likely to believe that something so helpful will end up as a crutch rather than a tool of learning.”

Anderson is adamant that students consider the long-term consequences when they’re thinking about cheating on an assignment. “We’re entering a period of history where our ability to think carefully and critically, precisely, and originally may be more important than it ever has been,” he said.

This story was originally published in the March issue of The Rubicon.