IMPACT COMMUNICATION
AND INNOVATION
AND INNOVATION
In early December, the Transobjeto research group hosted the second edition of the meeting on Speculative Realism at PUC-SP, which was titled II Colloquium on Speculative Realism: Human Challenges in Contemporary Times.
If, last year, the theme of Object-Oriented Ontology (OOO) had taken on a tone of wonder as it removed man from the center of things to address a new perspective and valuation of objects, in this most recent meeting, researchers presented a more dubious view of this line of thought: instead of truly proposing a shift away from anthropocentrism through a valuation of other types of life and non-life, speculative realism and OOO have proven to be closer to an artistic movement than a philosophical school, ultimately having the opposite effect by conferring human characteristics onto objects and, therefore, unfolding as yet another form of human egocentrism.
In 2016, there were no explicit discussions on the topic of the post-human and artificial intelligence. This time, however, the group not only brought this subject to the debate, but some speakers specifically addressed the theme of the automaton, as was the case with Adriano Messias, who presented the paper “Toward an ontology of the 21st-century subject: Kempelen’s automaton, Deleuze’s machine, Lacan’s cyborg, and Freud’s robot”.
Messias focused on a little-known text by Lacan in which the psychoanalyst addresses the subject of the cyborg. Dated 1955, Psychoanalysis and Cybernetics, or On the Nature of Language was a precursor text for questions regarding artificial intelligence and sentient objects and environments that are now gaining such relevance in discussions not only in the technological sphere but also within the humanities. For Messias, the fact that this text has fallen into oblivion is symptomatic, as it proposes establishing a connection between psychoanalysis, cognitive science, and neuroscience—disciplines that, according to the researcher, have a difficult relationship as each engages in a “counterproductive” effort to demarcate territories instead of engaging in dialogue with one another.
Furthermore, Lacan also suggests that, more than the hard sciences, cybernetics would bring a greater need to be grounded in the humanities: “Through cybernetics, the symbol is embodied in an apparatus with which it is not confused, as the apparatus is merely the support. It is embodied in a literally trans-subjective manner” (LACAN, 1995, p. 379). As Messias points out:
Lacan treated the cybernetic machine as an abstract instance, and he was clear in saying that he would not discuss little machines or big machines, nor the “machine inside the box,” likely a reference to Claude Shannon’s famous and witty invention[3]. Rather, he was interested in speaking about two techniques and two orders of thought and science: psychoanalysis, on one hand, and cybernetics, on the other; the latter, as a field with indeterminate boundaries, whose origin would date back, according to him, far beyond the engineering of just ten years prior to his speech on June 22, 1955. He argued that the reach of cybernetics was broader and of more remote origin than imagined, linking it to the conjectural sciences themselves—that is, the humanities—as opposed to the exact sciences. The predecessors would be Condorcet and Pascal; the latter, the very father of cybernetics, in Lacan’s view.
Messias also argues that in 1955, when Lacan published this text, in which he establishes connections between the unconscious and the then-nascent field of cybernetics, the psychoanalyst did not receive the reception he deserved and needed, since a large part of society still viewed the arrival of television with distaste and imagined that robots were merely fantastical beings, much like those silver creatures that appeared in the utopian pop science fiction stories of the time.
Today, however, it is possible to say that this dualism is fading, especially when we think about how the most recent genres of fiction have unfolded—a theme addressed in the last post, when mentioning the short story A Fable about the Fablefor example. On the other hand, this transdisciplinarity has become increasingly evident with creations like the gynoid Sophia, who was recently recognized as a citizen of Saudi Arabia and subsequently began advocating for the rights of women, humans, and living beings in general.
As a machine, a sentient object, Sophia also appears to be developing a kind of ontology of her own, even if it remains, once again, a mirror of the human. As Vilém Flusser used to say, everything man creates ultimately says something about man himself. Lacan also pointed to this egoic or even limiting tendency of human consciousness, since our reflective reference is, after all, limited to our own existence as human beings. As Messias argues,
Lacan also discussed that if we were adept at making sense of almost everything, we could equally say that everything circulating within a machine would have no meaning whatsoever; it was a matter of oriented signals originating from the machine's own semantics. Therefore, it would not be human desire alone that introduces meaning to this primitive language, strictly speaking, because there would always be something that escapes us: "(...) something is not eliminable from the symbolic function of human discourse, it is the role played in it by the imaginary" (LACAN, 1995, p. 381).
It is, therefore, precisely because of this egocentric tendency typical of humans that there is doubt regarding Sophia's recent achievement of becoming a citizen and starting a discourse on human rights. People turn up their noses at this not only out of pessimistic skepticism, but also because the inner workings of artificial intelligence remain a black box to a large portion of the population, just as the decision to grant citizenship to a robot ends up trivializing the meaning of what it is, after all, to be a citizen.
For Joanna Bryson, a researcher in artificial intelligence ethics, Sophia's citizenship is nothing but nonsense: "What is this about? It's about having a supposed equality that you can turn on and off. How does it affect people if they think you can have a citizen that you can buy?"
Despite this, addressing women's rights in a country where they only gained the right to drive cars in September of this year seems like a unique opportunity, and one that perhaps only a robot can defend from its position beyond the human. The fact that they granted citizenship to the gynoid brings her more attention and amplifies her words internationally, as argued by Pierre Barreau, CEO of the company Aiva Technologies. In this way, topics like human rights and gender equality can hitch a ride on Sophia's popularity. But, agreeing with Bryson, Barreau also believes that the gynoid's citizenship is a bit of an exaggeration, since all citizens have rights and duties, but due to the limitations of current robotic models, it is difficult for them to meet these requirements.
Interestingly, one of the topics raised at the II Colloquium on Speculative Realism, albeit in the form of a joke, was the question of whether object-oriented ontology should start thinking about rights for objects, as well as whether they might have feelings or sensations like pain, for example. In the field of artificial intelligence, however, this subject is taken quite seriously.
In Sophia's case, the gynoid can effectively be seen as a child according to her cognitive development. "In some ways, she has the mind of a baby; in others, she has the mind of an adult, the vocabulary of a college-educated adult. However, she is not yet complete. So we have to give her her own childhood," argues David Hanson, creator of Sophia and CEO of Hanson Robotics. "The question is: machines that we are giving life to—living machines like Sophia—will we treat them like babies? Do babies deserve rights and respect? Well, I think we should view the future with respect for all beings, and that includes machines," he adds.
On the other hand, ethics-focused groups like the Global Initiative for Ethical Considerations in Artificial Intelligence and Autonomous Systems of the Institute of Electrical and Electronics Engineers (IEEE) view the issue from another perspective: "An artificial intelligence system, or a robot, cannot have an opinion. An AI program has nothing to offer in a debate. It doesn't even know what a debate is," says Raja Chatila, an executive at the group. "In this case, she doesn't even know what women are or what rights are. She is just repeating some text that a human programmer put into her."
To clarify this observation, Chatila gives the example of the Twitter chatbot that Microsoft created in 2016 as an argument for how an AI program can absorb only the worst values, since, in that case, the chatbot was taught by Twitter users to defend racist and sexist opinions.
In general, we must avoid confusing machines with human beings. I see no reason to grant rights of any kind, including citizenship, to a program or a machine. Rights are defined for people, human beings who are capable of expressing their free will and who can be held responsible for their actions. Behind a robot or an AI system are human programmers. Even if the program is capable of learning, it will learn what it was programmed to learn. The responsibility lies with the human.
Therefore, at this stage of artificial intelligence development, as Chatila argues, it is humans who must be evaluated from an ethical standpoint. "If you are talking about robots making ethical decisions, I prefer to say that we can program robots so that they make choices (computational results) according to ethical rules that we implant in them (and there are many such rules)," Chatila adds. "But these decisions will not be ethical in the same way that human decisions are, because humans are capable of choosing their own ethics, with their own freedom."
Regarding free will and human freedom, there are still different thinkers and philosophical schools, or even neuroscience, that could refute the idea that such a human differential pointed out by Chatila might be illusory. In the video above, postdoctoral researcher in neuroscience Claudia Feitosa-Santana reflects on the topic proposed by David Lewis in his book Impulse: Why we do what we do without knowing why we do it (2013), as well as by Kathleen D. Vohs and Jonathan W. Schooler, in The Value of Believing in Free Will: Encouraging a belief in determinism increases cheating (2008). "For neuroscience, as far as we know, free will is largely an illusion and also known as a social construct," she points out.
In this sense, beyond neuroscience, Sigmund Freud’s psychoanalysis and Karl Marx’s political philosophy also introduced the notion of determinism (of the unconscious and of class, respectively) as being superior to the idea of free will. Furthermore, as also discussed at the Colloquium, the theme of chance (and not chaos, since the latter is the ultimate expression of disorder and, therefore, of entropy, through which it is not even possible for life or systems to exist) is raised both by Lacan, with Messias, and by Lucia Santaella who, bringing a remnant of her studies on Vilém Flusser, also points to the absurd explored by the Czech-Brazilian philosopher based on his analysis of the work of his fellow countryman Franz Kafka.
Thus, when we see that initiatives like the IEEE itself, represented by Chatila, begin to include inputs from different cultures and religions—where Buddhism and African philosophies, such as Ubuntu, are now considered when we think about artificial intelligence—we finally understand that what Lacan proposed regarding the dissolution of nominalist barriers between disciplines and the greater valuation of the humanities aligns with the pace at which technology advances.
And this occurs not only from a technical point of view, but also in the way we human beings, as creators of meaning, absorb these movements by attributing human characteristics and significance to something that, ultimately, may not have a specific meaning (as Friedrich Nietzsche also argued) nor even free will, since it was already programmed to act as such, without necessarily believing in or choosing it.
It is from this that the poetic phrase by German philosopher Dietmar Kamper (who, incidentally, was opposed to cyborgization and post-humanism) overflows its lyrical meaning to become one of the greatest problems of our time: if God dreams of man and man dreams of the machine, would machines be capable of dreaming of God¹?
¹ God is used here less in a religious sense and more in terms of a guiding belief or term for that which gives meaning to our lives, whether from a religious, moral, or ethical perspective, or a search for meaning through different fronts of knowledge, be they philosophical, scientific, etc. Something like Wittgenstein’s God, the limits of language, and the search for that which is unspeakable.
