Education, ChatGPT and the luddites of the third millennium

Online article Mondo Formazione by Massimo Berlingozzi, April 2023: https://www.hbritalia.it/mondo-formazione/2023/04/12/news/formazione-chatgpt-e-i-luddisti-del-terzo-millennio-15498/

It may seem like something very distant to evoke an event we studied in school – the workers’ protest movement that destroyed the mechanical looms accused of being a threat to the survival of labour – with the six-month moratorium demanded by more than a thousand signatories to the development of ChatGPT, OpenAI’s now famous artificial intelligence designed to answer questions by producing texts in a colloquial manner. In fact, a thin red line links that ancient protest with today’s call for a halt to the development of ChatGPT, not only because of similar concerns about the future of work, but precisely because Charles Babbage’s analytical machine, recognised as the forerunner of today’s computers, was born from Jacquard punch cards.

But here the similarity between the two stories stops, because ChatGPT does not merely threaten the jobs currently occupied by flesh and blood workers, it disquiets and frightens because it dominates the language that is the highest expression of our intelligence, and already drives the professors of universities halfway around the world crazy when faced with students capable of churning out well-written essays in the blink of an eye. So what to do, ban or govern this instrument? Faced with such fear, only careful reflection can help us make some essential distinctions.

First point: some fears are certainly legitimate, but a six-month moratorium will certainly not halt the progress of technology; some of the greatest philosophers of the 20th century have taught us: what is possible technology sooner or later achieves. Technique is a particular form of rationality typical of modernity, no longer limited to what we are used to calling the West of the world, so that it is in fact practically impossible to think of controlling what goes on in the thousands of research centres or even simple technology start-ups scattered in the four corners of the globe. But going beyond general considerations regarding developments in techno-scientific thinking, it is necessary to return to the question of what frightens ChatGPT in particular, and the answer can only concern its ability to give articulate, complex, syntactically correct answers within a conversation, as, indeed better than, many of us would be capable of doing.

The fear, and at the same time the very strong attraction, generated by being mirrored in an intelligence apparently similar to our own: this is probably the main key to understanding some of the motivations that have produced, at least in some of the signatories, the call for a moratorium. An appeal which, moreover, curiously sees the involvement of people directly linked to the development and financing of this instrument, so much so that one wonders whether it is not, in this case, a subtle marketing strategy. In reality, there is no intelligence like ours to worry about, but although this should be totally clear by now, for some strange reason many prefer to continue wallowing in this fanciful deception.

The result that ChatGPT returns to our questions is truly extraordinary, but behind those words there is no understanding, no awareness, just a huge statistical computing process capable of aggregating correlations of meaning in an endless sea of information with which that machine has been instructed. Inside the machine is darkness, as Federico Faggin, who designed and built those machines, states with an effective metaphor. This is the fundamental point that, however well known, should be continuously repeated and taught in schools to generations of children who will have to know how to work with these machines, being aware of what their role should be. The founding aspect of an epistemology that will have to guide the knowledge of the future, and this is where education comes in. If we do not understand that the profound and radicaldifference between the processing of human thought and the apparent processing of machine thought lies in the process that generated it, we are lost. For it is within the process of thought construction that lies the whole human quality, our history, our culture, our whole evolutionary journey. If we are not able to value this, we will soon begin to resemble machines, we will be training and no longer education, which has always been the place to exercise critical thinking for the development of knowledge and our awareness.

All this, after all, was already contained in the insights of some great thinkers long before these concerns were on the agenda. Thomas Eliot wrote in one of his poems in 1934: ‘Where is the wisdom we have lost in knowledge, where is the knowledge we have lost of information’, it is easy then to go back to ChatGPT and detect an immense amount of information without any understanding and therefore knowledge, no wisdom for lack of an ethic capable of questioning the meaning of one’s actions. And finally, Albert Einstein who said: ‘One day machines will be able to solve all problems, but never one of them will be able to pose one’. Never has it become more important than today to be able to ask the right questions.