ChatGPT Poses Unique Problems (and Solutions) in a School Setting

Even though it is relatively new, ChatGPT (and other AI) has started to change Barstow, and other schools around the world. 

OpenAI founder Sam Altman says he can imagine ways that ChatGPT 'breaks capitalism'
One of OpenAI’s founders, Sam Altman, talked to Forbes in February about his lofty plans for ChatGPT. (Steve Jennings/Getty Images)

Perhaps the most controversial tech innovations in the past few years, ChatGPT and other AI based chatbots have caused widespread uproar in educational settings. Generative artificial intelligence has caused many universities and high schools to revamp and restructure their homework options. 

In an interview with the New York Times, Northern Michigan University teacher Antony Aumann explained his experiences with ChatGPT, saying, “What’s happening in class is no longer going to be, ‘Here are some questions — let’s talk about it between us human beings,’” he said, but instead, “it’s like, ‘What also does this alien robot think?’”

To curb student use of the AI, teachers have turned to in-class assignments instead of the usual take-home assignments that have become popular in recent years. But at Barstow, students don’t really see a change. 

“My schoolwork has really been the same from before the AI was released to now. I still have essays to write at home, but I think the software will change a few things next year,” says junior Izzy Parr ‘24. 

For example, in Ms. Ganter’s AP Language course, students still wrote three take-home essays since the generative AI went live. Of course, essays are checked by plagiarism software (made by the company Turnitin), but ChatGPT’s capabilities often seep through these rudimentary checkers. 

The Barstow faculty, though, have been briefed by the Technology Department about the dangers of such automation. Additionally, OpenAI, the company that created ChatGPT, has released an “AI classifier for indicating AI-written text,” hoping to address concerns related to the chatbot’s misuse. 

The classifier is not 100% reliable, only identifying about 26% of AI-written work correctly. The software also works best on longer text, meaning that shorter Ai-generated text has a larger chance of not being detected. 

Another way to look at ChatGPT, though, is its positives for teachers and students in the classroom. For example, the AI could easily be used to create worksheets, assignments, and test problems to ease our faculty member’s heavy load. 

The chatbot also allows students instantaneous help, especially when stuck on subjects like math and science. In the case of AP Computer Science Principles, for example, students would be able to use ChatGPT or other generative AI to check and restructure their code for human error. 

To see ChatGPT in action, B-line tested its capabilities in writing a news article about itself, feeding the chatbot the prompt: “Write a journalistic article about the positives and negatives of using ChatGPT and other generative artificial intelligence in academic settings, like high school or college. Write the article in 500 words or less.”

Below are the (kind of terrifying) results: 

As technology continues to advance, artificial intelligence has become increasingly prevalent in academic settings. One area of AI that has seen a rise in usage is generative language models, such as ChatGPT. While there are certainly benefits to using these models, there are also some potential drawbacks that need to be considered.

One major advantage of using generative language models like ChatGPT is their ability to assist students in their research and writing. These models can help students generate ideas, suggest sources, and even help with the writing process itself. For example, a student struggling to come up with a topic for their research paper could input some keywords into ChatGPT and receive a list of potential topics to explore.

Another positive aspect of generative language models is their ability to help students who may struggle with language barriers. By using a model like ChatGPT, students can input their thoughts and ideas in their native language and receive assistance with translating and formulating those ideas in a more coherent manner in English.

However, there are also some potential negatives to consider when it comes to using generative language models in academic settings. One major concern is the potential for plagiarism. If students rely too heavily on the output of these models, they may inadvertently end up using language or ideas that are not their own, leading to accusations of academic dishonesty.

Another potential issue is the accuracy and reliability of the information generated by these models. While ChatGPT and similar models are trained on vast amounts of data, they are not infallible and may occasionally produce inaccurate or unreliable information. If students rely too heavily on this information, they may end up incorporating false or misleading ideas into their work.

Finally, there is the question of whether or not using generative language models in academic settings is ethical. Some argue that relying on AI to assist with research and writing takes away from the human element of these activities, and that it can even lead to a decrease in critical thinking and creativity. Others argue that AI can help level the playing field for students who may not have access to the same resources or opportunities as others.

In the end, the use of generative language models like ChatGPT in academic settings is a complex issue with both positive and negative aspects to consider. While these models can be a helpful tool for students, they must be used responsibly and in conjunction with other resources to ensure that academic integrity is maintained. As technology continues to evolve, it will be important for educators and institutions to carefully consider the role that AI should play in the academic world.

Scary, right? The new tech has the potential to revolutionize education, making it more accessible and dependable. But, trying to balance the human with the robot will pose myriad problems in the future. 

To try ChatGPT click this link, and to try OpenAI’s AI classifier, click here. . 

Author

  • Avani Lakkireddy '24

    Avani has been writing for B-line since 9th grade. Now, as a senior editor, Avani hopes to continue creating great articles, including focusing on more investigative work. Currently, Avani loves sports writing (specifically, soccer), as well as music and art reviews. She hopes you continue reading B-line!

    View all posts
The B-Line Staff thanks you for reading!

Never miss a post!

You'll only be updated when we post something new.

NEVER MISS A POST!

We only email when we have something new to share!