Interlude: The Pope and AI
On the Holy Father's address to the G7, "Mrs. Davis", and an idea from Lumen Fidei
In recent years, we have become used to grandiose pronouncements about the future of artificial intelligence. The line between the projections of futurist enthusiasts like Ray Kurzweil and out-and-out science fiction has become thin indeed. Yet when Pope Francis addressed the G7 leaders last Friday, he did not concern himself with warning of prospects of AI exceeding human intelligence, dreams or nightmares of conscious AI (with all the requisite suite of ethical obligations entailed), or condemning aspirations toward creating God. On the contrary, he offered a clearly delineated vision:
[A]rtificial intelligence is above all else a tool. And it goes without saying that the benefits or harm it will bring will depend on its use.
This is surely the case, for it has been this way with every tool fashioned by human beings since the dawn of time.
In a certain sense, AI is nothing special. Making tools has always been a part of being human; Francis calls this the “techno-human condition” and identifies it as “a sign of our orientation towards the future.” Yet having an orientation towards the future says nothing about the nature of that orientation – whether or not it is unto the common good and human flourishing. Naturally enough, this implies ethical and moral guidelines – no surprise again, for anyone aware of the Rome Call launched from the Vatican in 2020 and signed (for whatever it proves to be worth) by virtually all of the major players in Big Tech. For any of the ethical imperatives enumerated in the Rome Call to be realized, a values analysis of the issues specific to AI and to contemporary advanced technologies in general is necessary – particularly inasmuch as AI is a tool specifically designed to escape our control, or at least to operate outside of our control.
A significant point is that AI is a tool that makes choices – which shows more about the poverty of mere choice qua choice, than about any exalted, humanlike status for AI. After all, every day throughout the world, far lower-tech apparatuses choose winning lottery numbers, and no one accuses them of intelligence; in fact, it is precisely their lack of intelligence that keeps them immune to bias, and any extraneous insertion of intelligence into the choice process constitutes fraud. While equally material and formalizable, the algorithms and choice matrices involved in AI are so much more complex that they can pass for more than what they are, especially among those who consider human behavior to be itself material and (at least in principle) formalizable.
The Pope would not have us be satisfied with choice. As a Jesuit, he often expresses himself in terms of discernment; here, to a secular audience, he speaks of decision – which, he says, involves phronesis (practical wisdom) and “at least in part, the wisdom of Scripture.” Algorithms cannot discern or decide; their choices are at best expressions of systematizations – statistical summations or mathematical expressions of past human decisions or of trends in human decision-making. In all matters that concern human life, actual decision based on phronesis and discernment is necessary; anything less is essentially rolling the dice, however much we choose to load them to replicate our history of decision-making or our preconceptions about our potential future discernments. This applies equally to generative AI, which according to the Holy Father is not truly generative, as its manner of processing reflects the interests and biases of its designers, while its data set, wholly reliant on a past record of human decisions and actions, is inherently biased in favor of past choice processes.
All of this tends to “limit our worldview to realities expressible in numbers and enclosed in predetermined categories, thereby excluding the contribution of other forms of truth and imposing uniform anthropological, socio-economic and cultural models” – in addition to playing into the contemporary “technocratic paradigm” of tools and technologies so expensive, abstruse and lacking in transparency as to be exclusively the playthings (or worse, the weapons) of the wealthiest and most powerful individuals and nations.
One important point upon which the Holy Father touches comparatively lightly is the risk of AI becoming a substitute for human interaction:
[A]rtificial intelligence programs will be increasingly equipped with the capacity to interact directly (chatbots) with human beings, holding conversations and establishing close relationships with them. These interactions may end up being, more often than not, pleasant and reassuring, since these artificial intelligence programs will be designed to learn to respond, in a personalized way, to the physical and psychological needs of human beings.
It is a frequent and serious mistake to forget that artificial intelligence is not another human being, and that it cannot propose general principles. This error stems either from the profound need of human beings to find a stable form of companionship, or from a subconscious assumption, namely the assumption that observations obtained by means of a calculating mechanism are endowed with the qualities of unquestionable certainty and unquestionable universality.
This takes on a special interest in the light of earlier writing from Pope Francis (with a assist from Pope Benedict XVI) – specifically this, from Lumen Fidei:
Martin Buber once cited a definition of idolatry proposed by the rabbi of Kock: idolatry is “when a face addresses a face which is not a face”. In place of faith in God, it seems better to worship an idol, into whose face we can look directly and whose origin we know, because it is the work of our own hands…. Idols exist, we begin to see, as a pretext for setting ourselves at the center of reality and worshiping the work of our own hands.
Here I find myself turning to insights drawn from a woman religious and a prelate – namely the fictional Sister Simone Abbott of the (equally fictional) Our Lady of the Immaculate Valley monastery, and the quite real Most Rev. Robert Barron, ordinary of the Diocese of Winona, Minnesota. Bishop Barron was a fan of Sister Simone’s fictional milieu, the Peacock TV series Mrs. Davis. (I was not; I don’t watch television and elements in the Wikipedia plot summary put me off, rather badly.) Nonetheless, even a quick viewing of the trailer is enough to show that the character of Sister Simone is resisting what is in everyday, obvious terms an idolatrous cult; Bishop Barron stands at the ready to offer some theological underpinnings:
So pervasive is Mrs. Davis and so typically helpful that practically the entire human race has succumbed to her influence, gratefully doing her bidding and, with childlike affection, referring to her, depending on the country, as mother, mum, Madonna, and Mama. She has most of the qualities that one classically associates with God—virtual omnipotence, omnicompetence, and omniscience, even the capacity for providential guidance—and hence it is no surprise that nearly everyone reveres her.
But Simone has intuited that Mrs. Davis, in point of fact, robs people of their independence, saps them of their energy and creativity, controls them ruthlessly, and finally dispenses with them when they no longer suit her purpose. She has come to see, to state it bluntly and simply, that the algorithm is an idol, a pathetic simulacrum of the true God.
Yet I would add that what makes Mrs. Davis an idol and not a demon (pace Elon Musk) has nothing to do with any benevolent or malevolent traits she (“Not she. It.”) exhibits, but from the very trait the Pope cites in Lumen Fidei: that of being “a face that is not a face,” a not-real pseudo-person, a projection of our own notions, desires, and egos, a vehicle of self-worship.
As Val pointed out in our last post, faith is necessarily relational both laterally-ecclesially and vertically; moreover, Christian, Trinitarian faith is faith in an intrinsically relational God, a community of Persons. While the Holy Father loosely used the word “relationship” in his address with regard to Replika-style neural network machine learning avatars, he clearly recognizes the dangers in treating AI as if it were human.
One more positive I should acknowledge (even as a non-viewer) with regard to Mrs. Davis: the distilling of the kind of semi-mythical hype and solipsistic posturing that I discussed in the opening paragraph into arresting visual form. One example: the presentation of the earbuds in the trailer – note both the reverential, quasi-sacramental way in which they are offered, and the very fact that they are earbuds; this is an idol that does not wish to overawe, like the Wizard of Oz, but to ingratiate with inherently private and personalized access.
Of course, real-life AI is not so obviously cult-like (with rare exceptions) – but in certain ways it can become only the more insidious for not being so. In practical terms, this emphatically does not mean an anti-technological or anti-AI stance; rather it calls for what the Pope terms an “‘algor-ethics’, a series of principles [that] are condensed into a global and pluralistic platform that is capable of finding support from cultures, religions, international organizations and major corporations, which are key players in this development.” And this, in turn, he adds, requires political action and authority – a dirty word, the Pope acknowledges, for many in this time, but a word of truth, not to be neglected.
Such political action will require tearing down some pre-existing idols, especially in America: the idols of individualism, of market absolutism, of “that government governs best, that governs least.” It will also require tearing down the idol of political domination and the instilling of universal standards upholding the dignity of the human person. This, in turn, would demand wholesale conversion throughout humanity – at once the universal call of the Christian and the absolute last resort in terms of policy. That we may have no other hope is at once disheartening and oddly inspiring.
A final word: the response to the AI dilemmas mentioned above will require social trust (about which I hope to be writing more soon) – but, as you see below, social trust is under direct attack, not only from the individualism, contrarianism, and conspiratorial mindset we see all around ourselves, but also from AI itself. As you can see, “Pope Francis” can testify to that, as well:
.