Will Teachers Dream of Electric Sheep?

Written by sansoph | Published 2023/09/08
Tech Story Tags: ai-in-education | academics | learning-with-ai | chatgpt | classroom-technology | edutech | ai-writing | ai-in-classrooms

TLDRChatGPT has had people speculating about the nature and role of artificial intelligence in education and academia. However, these speculations often leave out real-world practicalities and the implications of AI integration into traditional learning mechanisms. This article traverses the numerous real-world possibilities of a potential AI adoption by academia. via the TL;DR App

A healthy imagination has always been a trait sported by human civilization. This potential for imagination is ever more magnified when it's projected toward conceptualizing an advanced future. Popular media of the Sci-Fi orientation only adds to this effect by making most of us imagine a world full of flying cars, android doctors, and interstellar travel.

While a lot of these imagined scenarios invariably play out to become starkly different realities, a small proportion are rooted in relatively tangible developments in the present. Among these is the innate human fear of replacement by machines and automated algorithms. While the anthropological origins of these anxieties might lay in humanity’s tribalistic past, it has come to embody economic and identitarian conflicts in the post-information age era of human society.

Among these, special emphasis is given to professions and skills that are considered innately human—either because of the element of critical thought, empathy, or intuitive decision-making approaches. Teaching and education are but some of the most identifiable with these exact attributes. However, with the arrival of the prodigal ChatGPT and a slew of other language model chatbots, the uniquely human claim to education has been put into question.

Although there has been a palpable increase in the investment to create newer, better, and coherent AI systems, the development of education-centric AIs has not quite picked up as much as one would have thought. So, are experts and academicians jumping the gun on this one? Maybe. While on one end doomsday predictions surrounding a complete AI takeover of the classroom abound, the other extreme seems far too dismissive and complacent about AI in education. The truth, like always, is somewhere in the middle.

Is AI in Education Even Practical?

The arrival of easily accessible generative AI opened up a Pandora's box. AI’s effects on the general populace—especially school and university students—took the world by storm in the late months of 2022 and in the opening weeks of 2023. Most news forums were busy discussing the rising usage of chatbots and other generative AI platforms to complete assignments and automate essential course-oriented exercises. This even prompted major universities in nations like France to ban LLM-based AI chatbots in early 2023. The ban was especially targeted at student misuse of ChatGPT and other generative algorithms for supplanting their efforts in academics.

Though this is evidently an ethical issue—one that’s been brought up rather frequently since AI presence began growing—it also presents a segue into the deeper aspects of integrating AI into an academic framework. While the ethical conundrums formed the bulwark of the pushback against the initial advances of the AI bubble, there also exist other nuanced concerns with the possibility of fully integrating AI with current academic infrastructures.

According to a 2022 report published by LinkedIn, premade AI solutions can cost anywhere between $5,000 to $50,000, if not more. Now, that’s a steep monthly bill for most schools and universities.

Most analyses often leave out the financial component of running, maintaining, and upgrading AI systems in an organizational format. Considering AI becomes important enough to academics that educational institutions actively begin considering dedicated AI deployment and support for their students, the financials become an inalienable part of the process. Since most educational institutions deploy anywhere between basic to intermediate tech frameworks and support systems, it’s safe to assume that these establishments might prefer subscribing to ready-made AI solutions.

According to a 2022 report published by LinkedIn, premade AI solutions can cost anywhere between $5,000 to $50,000, if not more. Now, that’s a steep monthly bill for most schools and universities. According to another cost breakdown for AI, the upkeep for maintaining a dedicated team catering to artificial intelligence needs might exceed $320,000 in just technology development, without taking into account overheads and other costs. Affordability, at least for the time being, is not a part of the recent AI boom and subsequent popularization. AI seems like a rather expensive affair even if academics actively consider it for their students.

Comparing Traditional Education and Artificial Intelligence

Apart from the depressing calculations from accounting, it’s equally important to assess whether AI is even a good fit for extant education systems. While the trend of associating AI’s adaptability with students’ idiosyncratic needs has become popular, it must be stated that most systemic changes are brought about by large groups and not individuals or smaller clusters of students. Traditional education approaches are by no means dynamic and humans have gone from using slate tiles to write on and learn during the dawn of civilization to submitting their assignments with a click on their laptops.

Although the methods and practices of traditional education systems might have transformed throughout human history, the central elements of human interaction, creativity, critical thinking, and independent reasoning have remained. Studies have shown that a creative learning environment can greatly enhance student outcomes and help build an overall innovative thought process in students. This innate characteristic of traditional education despite the percolation of modern digital technologies over the last three decades has remained more or less intact.

Regardless of speculative predictions, it has become increasingly evident that AI, too, is prone to its set of pitfalls and biases, and is limited by its determining factors like datasets and training protocols.

However, that might not necessarily be the case with the entry of artificial intelligence in the classroom and the larger academic landscape. While it may be true that AI can bring a refreshing departure from the one-size-fits-all approach of most curriculums with adaptive learning, any reduction in human interaction and a social environment might result in impacted learning outcomes. Now that AI has garnered a sort of critical mass with just about everyone talking about a looming AI takeover, better senses must prevail and assess compatibility before any major revamps to curriculums and educational frameworks are undertaken.

Regardless of speculative predictions, it has become increasingly evident that AI, too, is prone to its set of pitfalls and biases, and is limited by its determining factors like datasets and training protocols. Clearly, humans are still in the early stages of AI development, even by their own yardstick of tangible predictions about the domain. Any hasty integration of such mechanisms into the current academic syllabi might just kick up more pandemonium.

Will AI Take Over the Classroom?

Artificial intelligence in education is no new phenomenon. Its prevalence in rather mundane aspects such as administration and planning has already existed to a rudimentary capacity for some time now. However, technologies like AI-influenced student support services and learning solutions are still clearly a pipedream in the long and arduous process of developing capable artificial intelligence algorithms. Despite the uptick in the generalized usage of generative AI by students across different educational levels, this indulgence with a new piece of technology is more of an offshoot than the core purpose in and of itself.

Since these chatbots and their underlying language models are technological demonstrations, there clearly remains a greater scope that might take longer to actualize. Bearing in mind the negative short-term outcomes of learning sans human touch, socializing will remain central to the educational process regardless of whether or not we achieve superior AI capabilities. There’s no denying that AI will percolate into extant systems to cover more than the mundane, but the replacement of a human teacher might just remain a figment locked away in the many speculations of our collective human consciousness.



Written by sansoph | Sophia discovered teaching while juggling writing and a 9-5 desk job.
Published by HackerNoon on 2023/09/08