Me: What is your purpose?
ChatGPT: As an AI language model, my primary purpose is to assist users in generating human-like text based on the prompts or questions that I receive. I am trained to understand and respond to a wide range of inputs, and my responses are generated based on the patterns and information that I have learned from a large dataset of text.
In case you didn’t catch that, ChatGPT aspires to reinvent civilization as we know it by replacing our embarrassingly antiquated systems of critical thought. Artists, journalism, Google, ethics, morals…who needs ‘em?
One of the great privileges of my educational experience has been my access to a seemingly inexhaustible wealth of knowledge: Google. This seems to be a minor point of contention with the adults in my life, who purportedly spent years slaving away in the trenches of their school’s library shelves.
When hearing these stories I’ve often felt grateful that I never have, and likely never will, undergo such an exhausting venture. Sure, there’s a small sense of loss. Perhaps I’ve missed out on a major hallmark of the academia experience. But for the most part, I am appreciative of being born in an era where the libraries I depend on are digital, portable, and conveniently accessible at any time of day (or night). Indeed, my innocent high school self often marvelled at the ingenuity of the internet and could never have imagined a higher functioning power than the Google search engine.
But from preschoolers to university students, teachers to board directors, every member of the education hierarchy is about to become familiar with ChatGPT. The latest upgrade in our technological revolution can already code, answer multiple choice test questions, and compose English essays better than most high school students. It seems we’ve finally reached the holy grail of cheating.
The usage of AI and its implication on educational institutions will undoubtedly undergo the same type of scrutiny the calculator once went through. Progressives will argue ChatGPT has rendered many presently taught skills obsolete. And of these unstylish skills, my primary concern lies in the discontinuation of essay writing.
Specifically, I predict a new paradigm of writing pedagogy where emphasis will be placed on generating clear and concise directives to plug into your AI, followed by learning how to shepherd your AI to properly execute your ideas. And it isn’t entirely clear to me who should get credit for these compositions. A completely AI generated text is undeniably plagiarism, but if you feed ChatGPT the bones of the essay, does your name still deserve a place in the byline? The pejorative (and I myself), would answer no. But there is a case to be made that the AI-generated words are an extension of your thoughts, and therefore your intellectual property. (Perhaps we can compromise with ChatGPT & co.)
Naturally, the question arises as to whether a high school English education is even necessary to navigate ChatGPT. According to ChatGPT itself, apparently not.
Me: How much schooling would you require in a first world country to use ChatGPT?
ChatGPT: In a first-world country, the minimum level of education required to use ChatGPT would likely be the completion of primary or elementary school, as this would give you the basic reading and writing skills, as well as some basic understanding of how to use a computer and navigate the internet.
I would assert that the primary function of education is to lay a foundation of knowledge to teach students how to critically think, analyze, and execute their ideas in the real world. And while ChatGPT acknowledges we have yet to discard foundational necessities, I believe forgoing the practice of essay writing is a degradation of human intellect. Because more important than the essay are the thought processes used to guide its creation: language is an extension of thinking, and placing limitations on the way we employ language places limitations on human cognitive capabilities. Reliance on a tool like ChatGPT will breed stupidity, laziness, and apathy, and children who grow up under an education system that permits unchecked usage of AI will suffer the consequences.
The utopian-like quality of our present predicament seems to be the first step towards a future H.G. Wells predicted over one hundred years ago in his novel, The Time Machine. Some 800 000 years in the future, humanity lives in plenitude with all necessity for critical thinking having been eradicated by the ingenuity of their ancestors. Yet without the stressors and difficulties of primordial existence, they have cognitively degenerated into an infant-like species referred to as the Eloi.
Let me be clear: the threat of our species regressing into an Eloi-like state is improbable (or at the very least, distant). However, the illusion of absolute knowledge makes ChatGPT dangerous. Certainly, the PhD student who asks a question in their field can parse out fact from fiction. But the pregnable majority are not experts (which is why they ask in the first place) and are liable to consume ChatGPT as gospel. Unlike Google (which allows you to synthesise your argument based on the data available), ChatGPT resembles an intellectual God–benevolent, sagacious, and seemingly omniscient in its answers.
Me: Do you have bias?
ChatGPT: As an AI system, I do not have personal beliefs, emotions, or biases. However, the creators of AI systems like me assign ethical and moral principles that guide the behaviour of the AI.
In other words: yes.
Despite how it may claim otherwise, this AI system was modelled on human-generated text and would therefore have human-like biases integrated into its software. Thus, morals are assigned. Which in turn begs the question: Who gets to assign these ‘ethical and moral’ principles? You see, I find this statement quite ironic considering ChatGPT is about to become the #1 tool for academic dishonesty around the globe.
Leading up to the writing of this essay I’ve read a litany of articles complaining about the imperfections in ChatGPT’s system. So to those bemoaning that one erroneous answer it gave you: more important than the flaw you’ve discovered are the flaws you have not. This AI, even in its rough, test-trial version, marks a milestone in human progression. And as this technology continues to refine itself, its flaws will diminish and become increasingly less relevant. Do not fear AI for its limitations. Fear AI for its unrestrained ascension in competence.
I hope this software was created with the best of intentions. It will undoubtedly make certain aspects of our lives easier while making others more difficult. As you’ve probably guessed, I am a skeptic. I don’t believe an AI as advanced as ChatGPT should have been released without built-in, preemptive measures to prevent unethical usage.
I am also concerned with the lack of regulations on privacy. In the future if I were to search my name in ChatGPT, would it be able to composite my profile along with any accompanying information/pictures of me from the internet/social media? What happens to sensitive information I might leak to ChatGPT during conversations? (This is my plug to tell you NOT to leak sensitive information.) These are only some of the negative implications my measly brain could come up with, although I’m sure ChatGPT could generate a far more comprehensive list.
We are about to enter an ongoing arms race where our technological innovations are in competition with the laws we must conceive to regulate their usage. And ChatGPT is just the tip of the iceberg. Dall-E, GitHub Copilot, and DeepFake (a lawsuit just waiting to happen) are all on the rise.
So get ready.
Who is to say how AI will shape the future? Time will tell. But until then, I give the last words to ChatGPT.
Me: I understand there will be both negative and positive consequences for the increasing usage of AI systems such as yourself. In general however, do you believe ChatGPT will be a greater benefit or detriment to society?
ChatGPT:As an AI system, I am unable to have personal opinions or beliefs. However, AI systems like ChatGPT can bring benefits to society like automation of tasks, improved decision-making and valuable insights, while also raising potential ethical concerns such as privacy, bias, autonomy, and responsibility. It is important to approach the increasing use of AI with a balanced perspective, considering both benefits and negative consequences, and to address ethical concerns in a responsible manner.