Adopting ChatGPT into Our Everyday: What Users Should Keep in Mind

Adopting ChatGPT into Our Everyday: What Users Should Keep in Mind

ChatGPT is set to irrevocably change the way we write, especially as it begins to see widespread use throughout Singapore and the world. Whether as a student or educator, it is crucial to be aware of the uses and limitations of this conversational technology before adopting it.

Published On:

Author: Angelina Thng

 

 

The rise of conversational AI technology has threatened to significantly alter Singapore’s educational landscape. With the rapid adoption of ChatGPT amongst students, educators and educational institutions alike find themselves faced with the challenging task of spotting and mitigating abuse of the newly-prevalent language tool. In response, at least according to Education Minister Chan Chun Sing, schools and institutes of higher learning in Singapore have looked toward technological tools to detect AI-generated written content in assignments submitted by students.

 

Unlike many other countries which have come down heavily on the usage of ChatGPT in education, however, Singapore has instead advocated for an approach that integrates the software into society rather than rejecting what it views to be the inevitable future.

 

 

But what exactly is ChatGPT? A language model that functions in a dialogue format, ChatGPT interacts with the user conversationally. Research scientist Dr Kate Darling, who studies human-robot interaction, suggests that chatbots can be pre-programmed with prosocial responses that would encourage users to be nicer.

 

 

Nevertheless, this sort of question-and-answer format does not limit its capabilities to merely answering questions. For instance, it can be used as an aid for tasks such as crafting emails, writing essays, coding, creating apps, or even generally as a search engine that provides a singular answer. 

 

 

Additionally, ChatGPT can summarise content, describe art in detail, create prompts for art or writing, respond to philosophical questions and even help with putting together simple lists for everyday use. Even the Ministry of Education (MOE) has declared that resources and guidance will be provided so educators are able to make use of ChatGPT to enhance the learning process. Indeed, the AI tool is especially promising in terms of its ability to explain concepts or theories concisely, come up with teaching materials for educators and recommend references for further research.

 

 

That ChatGPT is capable of curating responses to the individual based on the questions they pose, and distilling complex ideas down to their essence is undoubtedly invaluable for the education sector. However, worries over students developing an over-reliance on such technological tools which would impede, critical thinking and prevent them from learning to write properly have been broached in Parliament. Reacting to this, Education Minister Chan Chun Sing uses the analogy of the calculator to explain the ministry’s approach to the new technology, implying that ChatGPT will be treated as a tool to support the learning process without necessarily removing the need for students to first master the fundamentals.

 

However, he has no qualms pointing out its limitations, stating, "As ChatGPT can provide inaccurate or biased output, students need to be discerning and critically assess its output for accuracy and objectivity." This point is extremely pertinent since unlike search engines, ChatGPT is unable to trawl the internet for the information the user seeks, instead drawing upon the data it’s been trained on to generate its responses. This results in the larger room for error the education minister speaks of, with the tool being known to make up information outright that is either false or inaccurate yet still plausible-sounding. 

 

Users should be aware that the quality of the language model’s responses can vary, oftentimes leaving much to be desired. At times, the answer may either come out overly verbose or more dangerously, sound reasonable without actually making practical sense. ChatGPT itself does not comprehend the meaning or verify the veracity of its statements and experiences difficulty distinguishing between falsehood and fact. It is also ultimately unable to take responsibility for what it says. This clearly imperfect technology essentially necessitates the work of a human being to edit and fact-check the output. 

 

Advice for Students and Educators Alike

While ChatGPT has the potential to be an extremely useful and accessible tool that improves our work and quality of life, it is better to avoid utilising its answers wholesale. Instead, treat it as a preliminary draft to edit, fact-check, and add personal touches to or as a tool to conduct your research. Be aware that answers generated by the chatbot can come off as mechanically polite and without human warmth, especially given that it is unable to produce metaphor, irony, or sarcasm in the way that a human writer would be able to. 

 

The large majority of educators view ChatGPT as a time and effort-saving supplementary add-on to bolster and streamline the educational process. In time, schools will begin teaching students how to best utilise and make the most of ChatGPT appropriately and safely without compromising intellectual property or putting oneself at risk of being accused of plagiarism. 

 

If you are a student, do note that some institutions do not regard AI-generated content as the student’s work and consider it a violation of academic integrity. When in doubt, approach your teacher or professor to ask about the school’s regulations as to whether the usage of ChatGPT or other such generative AI technologies is allowed. In using the tool, proper referencing and attribution is always necessary. Educators, too, should recognise that false positives are a possibility even with tools that claim to be able to register AI-generated content, with human-written work at times being wrongly classified as the work of AI. As a student, it is good practice in general to be able to show one’s proof of revision history, a feature which is readily available in both Google documents and Microsoft word.

 

Read more about how conversational AI might push back against toxic language and encourage people to do better, in BBC Earth Magazine (Asia Edition) Volume 15 Issue 3!

Topic:  DiscoveriesScience

Advertisement