Chat GPT and the Theatre of the Absurd

Though Chat GPT has been an important talking point in the world for some months now, it is only a few days ago that I signed up and tried my hand at it. I am rather slow with new innovations and technology. As I kept exploring what it does, at one stage I typed on the Chat GPT page, “Write a joke on Chat GPT.” As soon as I hit the enter key on my computer, it responded with this joke: “Why did Chat GPT refuse to play chess with the computer? Because it always thought ten moves ahead and got bored waiting for the computer to catch up!” Indeed, an interesting joke at the supposed slowness of computers and speed of Chat GPT. A joke after all is a joke, and no one takes it seriously as it obviously is a humorous take on something. However, I was rather impressed at the wit in the joke. 

Talking about the speed of both a computer and Chat GPT, the speed of a computer is incomparable with the speed of Chat GPT. The speed of a computer is dependent fundamentally on its hardware specification, the clock speed of its central processing unit (CPU) and its processing power. It entails how fast the computer performs its operations such as booting, opening programmes, processing data, and executing commands. Its speed is measured in terms of its clock speed and processing power. 

The speed of Chat GPT on the other hand entails how fast it can generate text-based responses against a given input. It is dependent largely on the architecture and complexity of the language model it uses as well as the hardware and software of the computer on which it runs. 

Again, I wrote on the Chat GPT page, “Compose a four-stanza poem on Chat GPT,” and hit the enter key. Exactly seven seconds later, I get the following poem. Since it tumbled out of Chat GPT without a title, I added the title, ‘The Digital Mentor.’

The Digital Mentor

Chat GPT, the marvel of our time,
With circuits built to think and rhyme,
Its artificial intelligence so bright,
Guides seekers through the depths of night.

From language syntax to deep debate,
Chat GPT’s responses never late,
Its algorithms trained with care,
A wealth of knowledge it does share.

Through wires and servers, it connects,
With those who seek to learn and reflect,
Its words of wisdom, sharp and clear,
A beacon of hope, free from fear.

A digital mentor, it stands tall,
With wisdom that can enthral,
A tireless teacher, day and night,
Chat GPT, a shining light.

I clicked on the ‘Regenerate response’ option on the Chat GPT page. I got another poem in a few seconds. Every time I clicked on the option, I got a new one. I copied the first poem, pasted it on an MS Word page, saved it with the file name, ‘The Digital Mentor,’ and uploaded it onto the plagiarism software, iThenticate, to put it through a plagiarism test. The test result showed 0% plagiarism. Despite not having written a single line, I can now posture myself as the supposed ‘author’ of ‘The Digital Mentor.’ Who can prove it otherwise? Chat GPT suddenly seems to make concepts such as originality, creativity, authorship, copyright, and intellectual property rights appear like redundant archaic concepts.

Also read: How to Grow and Care for hibiscus

I copied the text of the poem again and pasted it into Chat GPT asking this time for a critical analysis of the poem. The lightning-fast analysis commented on the poem’s conciseness, clarity, simplicity, celebratory tone, metaphors, rhyme scheme, enjambment, unidimensional language, and theme. It also commented on its over simplicity, formulaic structure, predictable pattern, lack of complexity, dearth of emotional depth, unilateral perspective (meaning that the poem focuses only on the positive aspects of the Chat GPT language model), failure to depict any potential negative consequences of artificial intelligence, inability to transcend the subject matter, failure both to connect with readers on a more profound level as well as to inspire them. The analysis concluded with the statement that though overall the poem “effectively captures the essence of the Chat GPT language model, it falls short in terms of its artistic merit.” Fantastic analysis!

Developed by OpenAI, an American artificial intelligence (AI) research laboratory, and released on 30 November 2022, Chat Generative Pre-Trained Transformer, known as Chat GPT, keeps the world in great suspense. Chat GPT, is a ground-breaking technology based on AI. It is an internet-based chatbot – an AI-based computer program designed to engage in real-time intelligent conversations with its human users. While Chat GPT is getting all the publicity, we must not forget that several other chatbots have been gaining popularity in recent times. All of them can simulate humanlike responses to inputs. The possibilities they offer are mind boggling.

Chat GPT, the most popular chatbot and our subject of discussion, can provide information as well as innovative perspectives on various issues, generate new ideas, create content, plan events, solve problems, and perform a lot more tasks as instructed. For example, it can explain what you ask it to explain, and compose what you ask it to compose, be it an article, a poem, a story, a film script, a speech, a summary of a text or even a translation of a text etc. It can provide instant answers to questions, suggest research topics, and provide feedback on written articles or assignments. As a compelling agent in imparting information and knowledge, it can customise and personalise the content and method of learning to suit each learner’s preference, pace, and style of learning. Surprisingly, it can reject inappropriate questions, dispute incorrect propositions, and even admit its own mistakes. The distinctiveness of all chatbots, also of Chat GPT, lies in their ability to generate real-time human-like responses to their user’s input.  

Developed by OpenAI, an American artificial intelligence (AI) research laboratory, and released on 30 November 2022, Chat Generative Pre-Trained Transformer, known as Chat GPT, keeps the world in great suspense. Chat GPT, is a ground-breaking technology based on AI. It is an internet-based chatbot – an AI-based computer program designed to engage in real-time intelligent conversations with its human users.

As part of an experiment, Prof Christian Terwiesch at the University of Pennsylvania’s Wharton School of Business, took an MBA exam at Wharton using Chat GPT, in order to see how the chatbot would perform in the exam. Interestingly, the performance was higher than average deserving B or B-minus level results. Impressive indeed! Discussing the experiment in his article, ‘Would Chat GPT3 Get a Wharton MBA,’ Terwiesch says, “Chat GPT3 does an amazing job at basic operations management and process analysis questions including those that are based on case studies. Not only are the answers correct, but the explanations are excellent.” However, “at times makes surprising mistakes in relatively simple calculations at the level of 6th grade Math. These mistakes can be massive in magnitude.” Therefore, he argues that “The present version of Chat GPT is not capable of handling more advanced process analysis questions, even when they are based on fairly standard templates.” Talking about the experiment and the possibilities of Chat GPT, Prof Terwiesch told Wharton Business Daily, a dynamic new morning show airing live from the Wharton School on SiriusXM Channel 132, “This is going to be big, and there is reason to believe we have only seen the beginning.”

The operation of Chat GPT is based on a type of AI known as the Large Language models (LLMs) that are designed to process as well as generate human language. The LLMs’ ability to handle human language is dependent on what is known as ‘machine-learning algorithms’ – a set of statistical and mathematical codes and methods which enable computers to learn structures, relationships and patterns of natural language based on large amounts of data available from the internet and other sources. Based on the learning from the data, the LLMs enable computers to come up with human-like predictions, decisions, and text reactions in response to a wide range of prompts. 

A search engine helps its user by searching the internet and indexing web pages containing the information asked for in order of the most relevant ones. Chat GPT, on the other hand, does not search the internet for information. Instead, using the information that LLMs have acquired through their exposure to and training on data, Chat GPT generates a single immediate response that does not require further perusal through different websites.   

As of now, LLMs have been trained on data only up to 2021. The more data the LLMs are exposed to and trained on, the better the Chat GPT’s responses in terms of logic, coherence, accuracy, and precision become. Since the world of the internet (a breeding ground of a lot of misinformation and incorrect data) is the major source of the data on which Chat GPT is trained, Chat GPT’s scope for coming up with erroneous and biased responses is a matter of concern.  

In its present form, Chat GPT does not have the ability to undertake research on its own. However, it can function as an efficient research assistant by helping the researcher in reviewing the literature, generating texts, analysing data, translating texts, and summarising written content. It can also let one get away with plagiarism, cheating and violation of intellectual property rights. Students are using it to do their homework, writers are using it to compose their content, and companies are using it to write their messages and perform other tasks. Despite all its advantages, concerned about the possible harm it can do to students and academia, schools, colleges, and universities in many countries have begun banning the use of Chat GPT on their institutional networks. However, there are many who argue that since AI is becoming an integral part of our future, it should be part of our learning process including the classroom. Perhaps it is time that Chat GPT entered our classrooms. 

With the eruption of the internet and social media over the last two decades, in general, most people’s reading habit has gone for a toss. Chat GPT takes the already-ruined scenario to further heights. Over-reliance on Chat GPT, especially by students and even academia, can paralyse their intellect – their ability to think, analyse and plan creatively and critically as well as solve problems and propose solutions. In other words, it can create people with A+ grades in borrowed artificial intelligence, the announcers of the doom of intellectual, academic, and professional excellence.

People across the world tend to turn to Chat GPT to find solutions for almost every problem they face. For example, many with mental health issues are turning to Chat GPT for counselling solutions even though Chat GPT itself put out a disclaimer saying that it is not a replacement for a counsellor or a therapist. However, people turn to it for solutions for their mental health issues because it can provide some sound advice, suggestions, and solutions that councillors and therapists often provide. Despite its possible drawbacks, pervading almost all aspects of human life, Chat GPT is taking on the role of the catholicon of our time.

With the emergence of chatbots, all search engines including Google, the most powerful search engine in the world, are likely to face an existential crisis. Matching chatbots’ ability to simulate superior responses and answers to the user’s queries will be the biggest challenge to all search engines. The possibility of job losses on account of employees getting increasingly replaced by AI is looming large as an absurd future scenario as depicted in the following joke I generated through Chat GPT: “Why did the company’s reliance on Chat GPT backfire? Because they forgot that even the smartest machine learning algorithms can’t fix the human errors of relying too heavily on technology!” As the joke suggests, chatbots, especially Chat GPT, has the potential to turn the world literally into a theatre of the absurd.

Images courtesy: Pixabay

Sacaria Joseph

Sacaria Joseph is an Assistant Professor in the Department of English at St. Xavier’s College, Kolkata. Having pursued his undergraduate studies at St. Xavier’s College, he furthered his academic journey by obtaining a Master of Arts degree in English Literature from Pune University, a Master of Philosophy from Jadavpur University, Kolkata, and a PhD from Visva-Bharati University, West Bengal. In addition to his academic pursuits, he writes on a wide array of subjects encompassing literature, philosophy, religion, culture, cinema, politics, and the environment.