Reflections on Early Reactions to ChatGPT in Education (Part 1 of 3)

#chatgpt is creating a plethora of knee-jerk responses in the world of #education and assessment, some unencumbered by a surfeit of knowledge.?This is the first of 3 articles in which I discuss what our early responses are telling us about its implications for educators and students.

Like most people, I experienced a variety of reactions and emotions as I saw ChatGPT in action for the first time. I’ve reflected on these as I’ve read through how other people have responded to it. In this first article, I explore whether the problems people are talking about are really as new as they appear.

REMEMBER ARTHUR C. CLARKE'S 3RD LAW

One of my favourite #technology quotes is apt here, especially if you’re coming to ChatGPT unfamiliar with the basics of how #artificialintelligence works conceptually:?

“Any sufficiently advanced technology is indistinguishable from magic”

Arthur C. Clarke, 1973

An Evolution of Familiar Technologies (Not Revolutionary New Technology)

My first reaction to ChatGPT was shock at what I seemed to be seeing. But this was closely followed by the realisation that, impressive as it undoubtedly is, it’s actually just a natural progression of familiar technologies we already take for granted: natural language processing, Internet search, and intelligent computer logic. Each is used in a very advanced way in the product, but arguably nothing its creator (OpenAI) has done is a radical technical breakthrough by itself - certainly compared to other advanced uses of #ai.

For example, looking at just the first of these, we’ve been literally chatting to Alexa and Siri for nearly a decade, Google Translate does an impressive job of understanding the meaning of words across languages, and Grammarly can ensure our writing is syntactically correct as we're typing. Hanson’s eerily humanoid robot Sophia has not only held her own in TV interviews, but was even given Saudi citizenship in 2017.?

None of these is perfect, but ChatGPT’s natural language technology doesn’t outshine any of them. The same goes for the other elements.

Massive Public Awareness of a “Free” Product (Remember Hotmail?)?

What OpenAI has done better than pretty much anyone is put these technologies together effectively. An equally impressive achievement is the dramatic public attention and reaction it’s garnered in just days. This is a result of another masterstroke - giving away ChatGPT free (at least for now). Again, not a new concept, but rarely used to such great effect. The last time was perhaps when everyone rushed to sign up for “free” email, even though there were plenty of other email providers around, some technically superior.

The important point is that the technology itself isn’t necessarily as big a deal as you might think. But even if it were, it would just be a matter of time before that changes. And with today’s technology, the timescales for such changes can be incredibly fast.

By the way, if it’s not clear to you why I feel as I do about ChatGPT’s technical underpinnings, that’ll be explained in another article, separately from this triptych.

HAVEN'T EDUCATORS BEEN SOMEWHERE LIKE THIS BEFORE?

My second reaction to ChatGPT - in the context of education - was that many of the comments seemed rather familiar. Others have shared the same view, but many talk about ChatGPT as if we’re entering completely unprecedented territory. I understand the sentiment, but don’t completely agree.

Dèjá Vu All Over Again?

A New York Times article about technology in schools reported that “Concerned father, Erik Sandberg-Diment was hesitant, fearing that [his daughter] would begin to lose her conceptual skills and become dependent on the machine”. But this was back in 1986, and was of course about the introduction of calculators into schools.?

In 1975, 72% of US teachers, mathematicians and the general public did not want calculators to be used in high school. Reasons cited included fear that children would forget basic skills, or even not learn those skills in the first place. As late as 1998, Readers Digest was still arguing that “[Students] often will not know if they even have the correct answer because they do not have a sense of what they are looking for.”

Research since shows that on the whole, most fears did not happen, that we found ways past problems that did arise, and the many benefits included ones we didn't (or couldn't) initially foresee. Today, the idea of education without calculators seems inconceivable.

This is a Really Old Debate

If you feel the analogy with calculators in teaching is too far removed from ChatGPT, there have been similar reactions over centuries to more comparable new technologies. Not digital of course, but affecting skills closer to those ChatGPT challenges.?

For example, printing and publishing was a new technology that became mainstream during the Reformation. Arguably, this was the first true Information Technology, literally technology to distribute information to many. Those in power were concerned about the dangers of this new technology on their people. Previously, the public had relied on approved, hand-written texts from their leaders - generally religious in nature - to tell them what they needed to know. Those leaders now feared publishing technology would lead to people’s morals being corrupted by this mass distribution of uncontrolled writing.

The Middle Ages were not the first time this happened. There are reports of such reactions to an even earlier form of communication technology: the papyrus scroll. Scholars have even suggested that Socrates - champion of verbal discourse and debate - warned of the negative consequences of writing for human memory and wisdom. Ironically, we only know this because it appears in one of Plato’s written texts.

What this first article highlights is that human fear of change is natural and perhaps inevitable, and technology can be powerful trigger for substantial change. That fear can be rooted in ignorance, or perhaps more accurately, uninformed reaction - especially while the technology is still new. Understanding the technology is important, both for a more informed response, but also to allow proper consideration of its impact on our lives and how we deal with that.

Education is of course hugely affected by all this, and is not immune to overreaction and under-informed debate. Perhaps most fundamentally, educators face decisions about how wider, technology-driven changes to life should be reflected in what we teach students.?

In the second part of this article, I’ll explore reactions to ChatGPT more specific to education. In particular, I’ll consider the concerns people are raising about the effects of ChatGPT in the exam hall and classroom, especially in the post-COVID world of Zoom calls and remote teaching. Once it’s available, I’ll add a link to it here.