Ai in education blog

Written on

Comment: The use of AI in education

Shelley Jones is a Graduate Teaching Associate in Aviation and Security, within BNU's Centre for Intelligence, Security and Resilience. She studied BSc (Hons) Intelligence, Analysis and Management and developed an interest in online investigations, especially around the utilisation of Open-source intelligence (OSINT) and ethical considerations. 

AI can be an incredibly useful tool in education to help students better understand complicated text, manage time, and paraphrase, however, should never be used to cheat or write the entirety of one’s coursework. In this blog, Graduate Teaching Associate in Aviation and Security, Shelley Jones, addresses the ethical and non-ethical ways to use AI, and explores that boundary, enabling students (and teachers) to become AI-literate.

ChatGPT is an AI chatbot created by OpenAI that was made publicly available in November 2022; since then, it has had a major impact on the digital and creative industry. Upon its release, ChatGPT was instantly regarded as a significant leap forward for conversational AI due to its ability to respond to questions and prompts with humanlike conversational dialogue. Similarly, DALL-E 3 (another product of OpenAI) allows users to generate images with prompts and is praised for its ability to interpret what kind of artistic output the user wants.  By January 2024, ChatGPT had amassed 100 million active users and fears were growing for online freelancers whose skills in copywriting and graphic design were now being replicated in seconds by the AI models (CNN, 2023).

In the case of education, AI models like ChatGPT pose a huge problem for assessing a student’s true understanding of a topic, and overreliance on this type of technology could suggest students are no longer retaining any taught knowledge. AI detection tools were quickly introduced into universities; however, these are never 100% accurate and there are a couple of ways to avoid detection, like using Quillbot (Another AI model) to reword the original AI text. Another loophole suggested involves copying and pasting AI text into ChatGPT and asking it to rewrite it in your style, which you would then paste into the command box for reference.

Another concern of ChatGPT and similar conversational AI models is misinformation. An article by The Guardian describes how a researcher had approached them regarding an article they had written on a specific subject, only this article did not exist. In response to being asked for articles on their research topic, ChatGPT generated references that appeared real but did not exist, a phenomenon referred to as an ‘AI hallucination’. This of course raises questions about the validity of ChatGPT outputs and the consequences for those who trust the model to aid in schoolwork or their professional lives. (The Guardian, 2023)

By April 2023, Italy had taken drastic action and banned ChatGPT due to privacy concerns over how ChatGPT processed users’ data and used it to train its models with little justification. ChatGPT was shortly reinstated later that year after the following changes were made: the addition of age verification to ensure users were above 13, increased transparency about how user data would be processed, and the introduction of opt-out rights that allow users to decline permission for their conversations to be used in the training of ChatGPT’s algorithms. (DW, 2023; BBC, 2023)

While we could argue for an absolute ban on ChatGPT in higher education, this is overall not an effective solution as students will still use ChatGPT outside of school and in their free time. If we allow the use of Grammarly, a very common and normalized AI tool used to assist in grammar and punctuation, surely, we can find a way to incorporate ChatGPT into the learning environment while still upholding academic integrity.

On that note, here are some suggestions on how to use AI ethically:

  • As a debate partner to generate arguments for your case or act as devil’s advocate to help you develop robust arguments against your opponent.
  • Ask the AI to summarize a large piece of text that may contain jargon.
  • Ask the AI to rephrase your work to meet the word count.
  • Generate discussion questions and outlines for assignments.
  • Generate feedback on assignments.
  • Generate ideas on how to update a curriculum.
  • To help you prepare for job interviews by asking for potential interview questions and how to answer them.
  • To make quizzes and exercises that target your current level of knowledge.
  • Translation of sources.
  • Analysing data sets. 

As shown by the above points, ChatGPT has the potential to be timesaving and reduce labour intensive tasks if used correctly. To excel in their careers, students must remain competitive, which inevitably will involve learning how to use AI and evaluating its quality and accuracy. I hope reading this short article serves as an important reminder that we need teachers and students to adapt to technological advancements, not feel threatened by them.

 

Reference List

BBC (2023) ChatGPT accessible again in Italy. Available at: https://www.bbc.co.uk/news/technology-65431914
CNN (2023) A year after ChatGPT’s release, the AI revolution is just beginning. Available at: https://edition.cnn.com/2023/11/30/tech/chatgpt-openai-revolution-one-year/index.html#:~:text=It%20wowed%20the%20public%20with,million%20active%20users%20by%20January.
DW (2023) Italy lifts ban on ChatGPT after data privacy improvements. Available at: https://www.dw.com/en/ai-italy-lifts-ban-on-chatgpt-after-data-privacy-improvements/a-65469742
The Guardian (2023) ChatGPT is making up fake Guardian articles. Here’s how we’re responding. Available at: https://www.theguardian.com/commentisfree/2023/apr/06/ai-chatgpt-guardi…