Chances are, a lot of you reading this article have an AI app on your phone. School work is hectic these days, and it’s way easier to pull up ChatGPT, search up a question, and get the right response in a matter of milliseconds. It correlates to our fast paced life.
We know that Artificial Intelligence (AI) already has a major influence on the lives of humans, but how far will it go? How is it affecting the ICSA community? And because of its endless possibilities, is it something we should be concerned about?
What is ChatGPT?
ChatGPT is an AI-powered chatbot that generates responses when given a prompt. It is powered by GPT (Generative Pretrained Transformer) architecture, specifically GPT-3.5. Did you know I used ChatGpt to write that sentence? All I did was type in a question and it gave me an immediate response. This is just a small glimpse into the power of this tool.
The Rise of ChatGPT
ChatGPT came out in November 2022, and it’s safe to say that the AI chatbot we all love and know now isn’t what it was before. Its first model, named GPT, was introduced in 2018—followed by updated versions GPT-2 in 2019, GPT-3 in 2020, and later on GPT-3.5 and GPT-4.
Shortly after its debut, ChatGPT went viral on social media platforms. Influencers were posting about how it could write blog posts, plan travel itineraries, and even write code for websites. Just five days after the launch, thanks to social media, over one million people had visited the ChatGPT website.
The first time I heard of ChatGPT was earlier this year through TikTok, and this is when I really started paying attention to it. When I scrolled, I saw people trying all types of things to test the limits of ChatGPT. An example of this is a Tiktok by NicolaiAI, in which he claims to have tricked ChatGPT. He did this by making the Android think that 10+2 is 15, when it is in fact 12. By simply telling the chatbot that it was incorrect, it responded with an apology, stating that it had made an error and that the user was correct.
Intrigued, I signed up for ChatGPT, as did many of my friends. And we were blown away by how much it could do. We’ve heard about students around the world using it to write essays and help with homework. To keep up, teachers have started to integrate the technology into their classes—using it to translate Shakespeare and to teach students about critical thinking. In fact, Open AI (the developer of ChatGPT) is creating a guide for teachers using the chatbot in their classrooms.
Is ChatGPT allowed to be used by students?
ChatGPT can help students prepare for tests, but it can also help students cheat on tests. This is not something that’s just happening here at our school. In fact, a study conducted in the UK states that 67% of secondary school students use AI for their homework and assignments, and 43% of US college students admit to using AI tools like ChatGPT for studying. By the way, that’s just the percentage of students that admit using it.
While many students will continue to use ChatGPT at home, many schools around the world are banning the technology due to plagiarism and accuracy concerns. And although there are no rules forbidding students at ICSA from using ChatGPT, it is banned on school-issued Chromebooks.
Good things AI is doing in education
At our school, many teachers are using some sort of AI to make their lives easier.
Even Mr. Pontius, Upper School Principal, uses it. “Yes, I use ChatGPT occasionally,” he says. “Sometimes to do some editing, sometimes to draft some first thinking to help get a ball rolling.”
He continues by saying, “Many teachers are using it. We had professional learning that specifically helped teachers lesson plan with AI to support learning engagements.” This could mean more fun and engaging classes for students.
In a recent Upper School survey, 75% of ICSA students use AI, and more than half of that percentage use ChatGPT specifically. School work was one of the main reasons for using it.
AI can provide teachers with feedback to help students with their learning. Some AI models can even act as practice students for teachers, going far as showing some levels of confusion. It is also good to note that some models won’t be doing the thinking for students anymore. Students will have to ask better questions, forcing themselves to think deeper than before.
Negatives about AI in education
One of the staggering limitations according to Forbes is that, believe it or not, its information is limited. It may not be able to answer some very complex questions. ChatGPT doesn’t have any emotional intelligence, so it might not be the best therapist to talk to. ChatGPT also lacks the basic common sense us humans possess.
“I am sure there are students who are relying on it too much, trying to get it to do their thinking for them.”
– ICSA Upper School Principal, Andy Pontius
Another one of ChatGPT’s flaws is that it has a biased system. This is because ChatGPT gets its information from a variety of human-created texts. As Mr. Pontius points out, “All of us need to learn how to use it responsibly. There is bias contained in AI, and we need to be able to identify that.”
Ways ICSA students use AI
While most surveyed students say they use AI for their school work, many mentioned other practical uses. An anonymous student stated that they use AI to help them interact with someone they like.
Another student stated that they use AI for translation. Another stated that they just like talking to robots and making random conversations with them.
With how things are looking like right now, is AI going to be teaching students and not actual human teachers? AI is an asset to humanity, but it is also a liability. We humans, especially students, need to be careful with how we use it.
“We need to still do creative thinking and be the ones behind it. It certainly can be a time saver on certain tasks, but the importance of learning before relying on it will be essential if students want to have success in the future,” says Mr. Pontius.
Humans might rely on AI from time to time, but at the end of the day, we have to rely on ourselves even more.