Focus on nexuses of complexity: Education

Being human in the time of neuroscience and artificial intelligence involve carefully exploring the nexuses of complexity where valid ideas are nevertheless in tension, manifesting subtleties and challenges that must not be overlooked. Each page expresses the existing tension(s) between ideas and within each theme, which emerged in the collective discussions, and are then complemented by insights from NHNAI network researchers.

The participants highlighted the benefits that AI can bring to education. Starting with digitization, which makes online teaching materials accessible to anyone, facilitating instruction outside class hours, enabling pupils and students to extend subjects seen in class, and making it easier to catch up on lessons when absent, thanks to online school platforms. Online discussion and debate forums also enable people who are too shy or less comfortable speaking to express themselves. AI presents itself as a virtual assistant that can help with language learning. AI-assisted language learning is becoming more accessible thanks to translation systems, which are now indispensable for people with language difficulties or for the deaf or hard-of-hearing, as mentioned in Kenya and France. And as language learning partly requires oral practice, conversational robots are sometimes more effective than language books. This is exactly what chatbots like ChatGPT can be used for. Used wisely, they can be a formidable pedagogical tool, a necessary aid to learning and complementary to the teacher. In addition, the complementary nature of AI and the teacher was emphasized several times in the discussions, and this is illustrated in particular in the personalization of learning. AI makes it possible to personalize learning paths according to each student’s pace, level and ability. As it is physically and cognitively impossible for the teacher to take into account the specificities of each student, AI enables him or her to have an overall view and to identify students in difficulty who are in greater need of support.

But participants also recognize that AI’s contribution to education (more inclusion, more access…) very often comes at the expense of physical interaction and human contact, and this concern was almost unanimous in the discussions. The availability of online learning materials can also have the negative effect of encouraging students to invest less time in classroom activities, or even prompting some to drop out and home-school, given that everything is now available online, and within everyone’s reach. In Portugal and other countries, there is also a risk evoked that younger people, having become accustomed to this new format of online relationships, will become content with these virtual contacts and underestimate their relational, emotional and physical needs, to the point of becoming distant and cold in contact with others. According to one participant, we can’t do without real face-to-face interactions when it comes to learning “how to be, how to know and how to act”. But beyond this learning, it’s also in physical interactions that empathy, emotion, mutual and reciprocal understanding – in short, the encounter with the other – come into play. And, as one participant in Canada pointed out, it is sometimes the presence of a teacher and the transmission of his or her passion and emotions that play an important role in the learner’s motivation and attention, and therefore in his or her learning. So school is not just a place for learning, but also a place for sharing, meeting new people, and learning to live together, to help society flourish. Through physical interaction, we confront each other, learn social codes and pass on values. Digitalized education, or education that takes place too much behind screens, can ultimately run the risk of reinforcing individualism and egoism, which would be a major brake on living together and a threat to social cohesion.

Based on insights from Brian P. Green (professor in AI Ethics, Director of technology ethics at the Markkula Center for Applied Ethics (Santa Clara University, USA) and Laura Di Rollo (research engineer in cognitive sciences for NHNAI project (UCLy (Lyon Catholic University), UR CONFLUENCE : Sciences et Humanités (EA 1598), Lyon, France)

First, what the participants in the discussions are expressing is a paradox that Sherry Turkle illustrates through the title of her book “Alone together” (2015),[1] with the concern that young people are no longer investing in human relationships, and that more is expected of technologies than of humans. And yet, education-focused relationships are among the most important relationships we have as humans. Most people, as the participants in the discussions also raise, can remember someone who taught them something, whether it was our parents, a friend, or a teacher in school. These relationships are vital to our humanity and AI puts them at risk, particularly in two ways: 1) as a distraction from learning, and 2) as a replacement for learning. AI which is optimized for grabbing attention distracts students from their work. And AI which generates text and other “education-product-like” objects permit students to replace themselves with AI in the learning process, thus learning significantly less.

Humans need each other, especially for education. Without socialization and education we become feral, like domesticated animals raised in the wild, but in the case of humanity, rude and vice-ridden. In order to become fully human beings, children should not be raised by screens and algorithms, but by other fully human beings.

In addition to the risk of disinvesting in human relationships, digital technologies also present the risk of commodifying the human being, i.e. reducing him or her to an object. Indeed, as Sherry Turkle (2015) points out, the risk is that our “self” is transformed into an online “object-self,” where we treat each other more and more like objects and in an expeditious manner. The most telling example is certainly email. Emails are a cognitive load in themselves, but sometimes they’re messages from friends or colleagues that we say we need to “deal with” or get rid of so we can cross them off our to-do list, as if we were talking about our dustbins.

Ultimately, the danger is that we lose the feeling of being alive, the way of being-in-the-world that preserves a certain dignity and authenticity, and that only human relationships and face-to-face contact can provide. AI has the potential to be a weapon of mass destruction upon the world’s educational system. It needs to be disarmed and instead harnessed as a source of power to assist humans to become better people rather than harm us by enabling the worst parts of our natures. Thus, it seems necessary to strike a balance so as to benefit from what AI can bring us, while preserving those precious human contacts that largely define our humanity, notably through certain attributes. The human voice is to Sherry Turkle what the face is to Levinas.[2] For Sherry Turkle, it is in the voice that the range of human emotions and the singularity of being are transmitted and heard. For Levinas, it is through the face that the other appears to me in his or her fragility, vulnerability and singularity, which calls for an ethical injunction to protect and not to harm. The face is an interface that enables us to enter into a relationship with others, and through them, with humanity. This raises the question of whether the danger threatening humanity, with relationships mostly at a distance and mostly faceless, is not indifference to the other, and with it, the loss of concern for humanity.

[1] Turkle, S. (2015). Seuls ensemble. De plus en plus de technologies de moins en moins de relations humaines. Échappée (L’).

[2] Lévinas, E. (1984). Ethique et infini. Le livre de poche

Brian P. Green (professor in AI Ethics, Director of technology ethics at the Markkula Center for Applied Ethics (Santa Clara University, USA)

AI as a driver of social and economic inequality is an inescapable question because AI will reduce the value of labor and increase the value of capital, thus driving wealth away from workers and towards owners of AI. How to prepare students today for the strange world of tomorrow, where labor might have no value and only those who already own wealth will retain wealth is an unsolved problem of gargantuan proportion.

Students need to know that a strange future is approaching and to be aware of AI and neuroscience as developing technologies that can affect their futures. Additionally, the uncertainty that will be sparked by these revelations should not be allowed to overpower the growing importance of particularly human pursuits such as seeking ethics, justice, and creating a more caring world. While intellectual labor might be, in some cases, automatable, caring relationships between families and friends can never be automated. Particular human relationships are not fungible and therefore AI can never replace them. The value of family and friends should be re-emphasized and the study of what makes good relationships should be a key part of the revision of education.