Curiosity

There is no doubt that AI is a technology like no other and that, as a result, the disruption it is likely to cause will be unprecedented. But despite the existential fear that this has brought about, it is unlikely that it will render us irrelevant. Not so long as we continue to constantly question.

This is a link-enhanced version of an article that first appeared in the Mint. You can read the original here. If you would like to receive these articles in your inbox every week please consider subscribing by clicking on this link.

Every time we have been exposed to new technologies, we have worried about how they will change our lives. And yet history has shown that we usually overestimate the disruption in the short term and underestimate it in the long term.

Today, we stand at the threshold of the artificial intelligence (AI) revolution. As before, worries about how this new technology will affect us—at work and play—are already the subject of debate and discussion. But if, like all other technologies that came before it, the immediate short-term disruption will eventually make way for benefits over the long term, should we simply bide our time?

Or is AI somehow different?

Technology Disruptions

When the mechanical loom decimated the handloom industry, it unleashed a textile boom that made clothing cheaper and more widely available. Even though the sewing machine ended bespoke hand-stitched tailoring, it democratised fashion and gave rise to ready-to-wear clothing. Electric streetlights put lamp lighters out of a job but also ended the tyranny of darkness, letting us pack more into our days. Electricity was a general-purpose technology that destroyed more traditional ways of working than anything that had come before it, creating new industries that couldn’t have existed before.

In offices, typewriters threatened traditional clerks and scribes, but were instrumental in creating a professional secretarial staff that played a role in bringing women into the workforce. Word processors and personal computers replaced typewriters, and in the process created new jobs in IT and desktop publishing, apart from a whole new culture of knowledge work. Spreadsheets didn’t replace accountants, as we feared, but instead took the drudgery out of repetitive calculation, allowing humans to focus on financial modelling and analysis.

AI is the world’s latest technological transformation and its effects are already evident. Some companies have already put a freeze on hiring, committing to replace employees with AI systems they believe will be more effective. As for the rest of us, AI is changing our daily lives through personalization and the rapid anthropomorphization of its interface. But while these first-order effects are evident, the long-term consequences of this technology remain unclear.

AI and Academia

It would, in this context, be instructive to study the impact of AI on academia. In a recent article in The New Yorker, humanities professor Graham Burnett articulated the discomfort he felt when, after a career producing knowledge, it became evident to him that AI systems could do so as well. What would remain for scholars like him, he asked, when a machine could do what he does—and do it better.

The core tenet of our academic system is that mastery over knowledge develops gradually through methodical exposure to academic concepts of incremental complexity. We implement this by systematically immersing students in information, compounding their knowledge gradually over time under expert guidance.

AI systems force us to question this approach. If students can use AI to get the information they need in a fraction of the time it would have taken them to find it in a library, what is the point of training them in traditional research methodology? If an AI chatbot can explain concepts to them in ways that are ideally suited to their own unique learning styles, is there any point in insisting on rigid pedagogical frameworks designed to cater to the broadest cross-section of students? When powerful reasoning AI models can generate detailed reports on just about any subject, does testing students’ knowledge through class assignments still make sense?

Our education system has always been about textual knowledge. Academic achievement is equated with the production of monographs and progress defined by knowledge of texts. But all that loses its relevance in a world where AI systems have ingested every last shred of textual information and compressed it into neural networks so vast that information can be summoned instantly. What is the relevance of text when every new paper that is produced is more likely to be summarized by AI than fully read, and when the number of papers generated by artificial systems exceeds those written by humans? And, more importantly, how should traditional academics adapt to this new world?

Curiosity

The answer to that question lies in what it really means to be human. As a species, we have always been curious. This is what has separated us from our animal ancestors and taken us to the top of the planet’s food chain. But it is not the answers we know that make us human, but the fact that no matter how much we have learnt, we cannot stop asking questions. Indeed, this is the essence of our humanity—this constant quest for new explanations, for no better reason than the fact that, as humans, we have a constant need to know. In the words of Dr Burnett, “..to be human is not to have answers. It is to have questions—and to live with them. The machines can’t do that for us. Not now, not ever.”

If AI feels unlike anything we have faced before, that’s because in many ways it is. But even with all the world’s knowledge at its fingertips, what it can’t do is ask questions. It is this insatiable curiosity, the impulse to always ask what lies beyond every answer, that makes humans special.

After all, even after we find “the answer to life, the universe, and everything,” our next instinct is to ask what the real question was.