Written By Rishikesh Mistry and James Inglis

AI, or artificial intelligence, has become increasingly prevalent in recent years, with advancements in the field leading to the development of new technologies that have the potential to transform the way we live and work. One such technology is ChatGPT, which uses AI to facilitate natural conversations with humans. It has been developed by Elon Musk’s team at OpenAI and is now one of many open access AI’s that help people to create content. In this article, we will explore the implications of ChatGPT in the education sector, specifically in schools and in Guildford. We will examine the potential benefits of using AI for students, teachers, and schools, as well as the potential challenges and ethical considerations that come with the use of AI in education. Ultimately, we aim to provide a balanced and nuanced perspective on the role of AI in the education sector and its potential impact on the future of learning.

People place too much trust in the ChatGPT system which opens up the opportunity for the proliferation of misinformation if ChatGPT answers incorrectly. OpenAI does provide disclaimers that there AI is not able to answer perfectly correctly every single time. It may provide answers referred to as a “hallucination” of fact in which a plausible-sounding answer is provided although it is in fact nonsense or incorrect. A particularly dangerous implication of this is in the context of medical advice where erroneous information could potentially lead to harm to people. The AI is not designed to provide this sort of information and so it is not able to perfectly provide this information however some people may be unaware of this. Information is not taken directly from the internet as it is provided with access to it and information published after 2021 is unlikely to be found. Therefore, asking for information which may have changed in an essential manner after this date could also lead to the wrong answer being provided to the user.

ChatGPT has demonstrated bias due to its dataset. When training the AI, the neural network must be provided with data which is originally from people and hence bias can transfer from data to AI. There have been occasions where the AI provided answers which contain prejudice against certain groups of people. This could be harmful as the user receiving this answer could take in the bias subconsciously and present it in their day-to-day life. To combat this, OpenAI has asked its users to provide feedback on responses to flag potential “biased behaviour” as it is referred to. Some people have criticized OpenAI for releasing the AI prior to reaching acceptable resolutions to this issue as other Companies such as Google and Meta have done with their language models.

ChatGPT presents a genuine threat in education, particularly pre-university as many students learn in classes such as English or other humanities the skill of writing well and producing coherent essays. However, the capability of the AI to proofread and improve the writing of a paragraph whilst maintaining its style presents the question of whether this skill will continue to matter. Students can also use the language model to create entirely original essays without having to create their own template paragraph to start. This creates the potential for many students to write essays throughout the school year using ChatGPT and then have never developed the ability to write for themselves which will prove important in the exams they take. OpenAI are however developing a method to detect the use of their own tools in the writing of an essay so schools could perhaps employ this to combat this effect. However, the question of whether such a tool is possible remains unanswered. Furthermore, some oppose the creation of such a tool with it being viewed as opposing progress and limiting the capabilities of ever developing technology.

ChatGPT is a highly sophisticated language model and possibly the most powerful one available to public free of charge. This allows it to be used by almost everyone extremely easily as it has a user-friendly interface and low barrier to entry with only the creation of an account required. It is able to understand a vast array of inputs and produce a very human-sounding response with often striking amounts of detail and information. Even in the case of inputs containing highly technical terminology from more obscure or esoteric fields, the AI is unable to interpret and “learn” in real-time if the terminology is defined to the AI. This provides people with the ability to do a form of research with specifically tailored responses on practically every subject available with only those at the cutting-edge of development being unlikely to be compatible due to ChatGPT’s information coming from pre-2021.

Using ChatGPT, a student could create a tailored learning experience in many fields from the comfort of their own home. The difficulty of self-teaching is often that there are no teachers or educators present for the student to ask specific questions to explain concepts or ideas that may not immediately make sense. ChatGPT has shown to provide accurate information on most topics most of the time and so a student could pose their specific question to the AI and have a response tailored around that given to them. This makes it easier for people to learn complex topics from home which improves overall education and may ease the pressure off of learning institutions such as school or university.

ChatGPT could further be used as a legal assistant as it would be able to pull information from a much larger range of case studies no matter how obscure they may be. This would certainly improve the consistency of the legal system as by referring to previous convictions or lack thereof, a ChatGPT legal assistant would ensure that the law is repeatedly applied in the same way. It could also be used to create persuasive writing which could be read out in court in order to convince a jury of a verdict. It should be mentioned however that there is a danger in criminals from organised crime using their vast resources to create incredibly strong legal defences using AI and avoiding conviction.

In terms of Guildford and the local community, I think we can hope to see increased productivity and: Chatgpt and other AI could potentially help small businesses automate certain tasks, freeing up time and resources for more important tasks. This could lead to increased productivity and efficiency, potentially giving small businesses a competitive advantage. On the other hand, the use of AI technologies like  could potentially lead to job displacement, as certain tasks and responsibilities may be automated.  The use of these developments also raises a number of ethical considerations, including issues related to privacy, bias, and accountability. It will be important for businesses and the town of Guildford to carefully consider these issues and take steps to address them. I think a quote that sums my point up really well from an RGS student is “business need to adapt quickly and utilise it [chatGPT] effectively, in a couple of years everyone will have caught on so it’s important to learn now”.

In conclusion, ChatGPT has the potential to revolutionize the field of education. Its ability to understand and respond to basic statements can provide students with personalized and engaging learning experiences. However, it is important to note that there are also potential drawbacks to using this technology, particularly the use of AI for learning and completing homework. Overall, it is important to carefully consider the benefits and limitations of AI before implementing it in the classroom, and to continually monitor and evaluate its effectiveness.