Listen to this story
After the launch of ChatGPT, every industry in the market is figuring out how to make the best out of this technology. Alongside this excitement, it has also caused concern among several startups and employees, as they fear that AI is looming over them, potentially taking away their jobs.
However, among all the AI buzz, there is one sector which appears to be carefree and is taking AI lightly without realising its consequences. It is none other than educational institutes. This approach might hurt them in the future if they don’t take necessary steps.
Whenever a new technology comes in, there are always early adopters and late majority, according to the Theory of Diffusion. In the case of foundational models and generative AI, no one had expected that it would be universities and schools which would be shying away from it while they should have been the first ones to integrate in their teaching methods.
As soon as they came to know about ChatGPT, the knee jerk reaction of the majority of the universities and colleges around the world was to ban it.
Why, just because they feared that students might use it to copy their assignments. This solution is surely not going to work in the long run. Students are smart enough to outwit their professors if they want to cheat in their assignments.
For instance, when Google Search came in, it didn’t stop students from surfing the internet for completing their assignments. To tackle the issue of plagiarism, several tools like Turnitin emerged. Similarly here professors need to come up with better solutions to handle AI. One thing is for sure, running away isn’t the solution.
Accept it or get doomed
The ongoing discussion among the AI industry revolves around the potential threat to the existence of educational institutions if they don’t adapt accordingly. Recently, a X user posted “We should decimate educational institutions with AI”.
Similarly, Ethan Mollick, professor at Wharton posted on X “The start of the school year is AI chaos, with many instructors just ignoring AI. I think instructors need to make active choices about AI use (which might mean embracing AI or returning to in-class tests) and there is near certainty that models will improve over the school year.”
Mollick is not alone in expressing these sentiments. “Educational institutions are completely unprepared for the AI revolution. They managed to wing it last school year, but I feel they will be completely overwhelmed this coming year,” posted Bojan Tunguz, machine learning scientist at NVIDIA, on X.
It’s pretty much evident that those who will embrace AI will survive and those who won’t are in for a hard time.
Rather than seeing AI as a threat, educators can view it as a valuable partner. AI-driven chatbots, for instance, can provide immediate responses to student inquiries, freeing up professors to focus on more complex aspects of teaching.
Echoing similar sentiments, Rajeev Kumar Singh, Associate Dean, Academics at Shiv Nadar University in a recent interaction told AIM that he is optimistic about adopting AI tools in teaching methods. He explained this will give faculty some extra time to spend on discussions with students one on one.
“Let’s say I take 45 lectures and I find that in 10, I was only transferring information. So, if that can be outsourced to some technology, I can spend the entire rest of the lectures on really bigger things. If I can save that time, then I can really engage one to one with students, which can be very fruitful, it can lead to a lot of personalised assessments, learning and also relationships.”
It’s understandable that it is tough for educators and professors to start from scratch to understand this new technology. However, it is for their betterment only. Otherwise chances are pretty much that AI will replace average teachers and amplify skilled educators. Earlier this year, Harvard University announced that its coding course, CS50 will be taught by an AI instructor. So, the reality isn’t far away.
To give themselves a head start, professors can go through a five part YouTube series on ‘Practical AI for Teachers & Students’ created by Mollick. Mollick went on to say that he agrees it is a lot for instructors to handle among their other obligations, “but the tech isn’t going away.”
To assist educators with AI, he has even published two papers ‘Assigning AI: Seven Approaches for Students, with Prompts’ and ‘Using AI to Implement Effective Teaching Strategies in Classrooms: Five Strategies, Including Prompts’.
Not only that OpenAI recently took initiative to educate teachers how they can use ChatGPT in schools and universities. They gave several examples on how teachers can come up with lesson plans made with the help of ChatGPT.
Moreover, universities themselves can organise sessions designed to educate both faculty and students about the implications of large language models by partnering with industry experts. For instance, Shiv Nadar University recently hosted one session titled “The New Landscape: Designing in the Times of Dall-E, Midjourney, and ChatGPT!”.
It’s not too late
The good news is that educational institutions still have time to make corrections. The process of integrating AI can begin with the admission process. Universities can consider developing their own LLMs, which would greatly assist students in navigating the admission process.
Similarly, schools and colleges can create a process that emphasises students’ critical thinking skills while still allowing them to utilize ChatGPT for assistance. There is no denying the fact that ChatGPT provides the best answers when asked the right questions.
Many believe that personalised AI assistants are going to be a thing in the future. If we extend that idea to education soon we might also get personalised AI tutors who would be teaching students according to their requirements and understanding capabilities.
This will solve Bloom’s Sigma 2 problem. Bloom’s research found that students who were taught one-on-one or in small groups and received regular feedback performed two standard deviations (2 sigma) better than their peers who received traditional classroom instruction.
One can get a rough idea of how the future is going to look from Musk’s private school he created for his children—called Ad Astra. It is quietly built inside the SpaceX campus and has partnered with Synthesis, founded by Joshua Dahn. Synthesis has created an AI Tutor which teaches students complex concepts of math through personalised games and AI tools.