A Quick Ramble: Computational Compressibility (order in the universe), Religion & Signaling, Silicon Valley vs. Pope Leo [+Latour]

I’m thinking there’s a relationship between computational compressibility (as an index of order in the universe) and my current interest in religion, which is what the other two items are about.

Computational Compressibility and order in the universe

I’ve been particularly pleased by my recent working paper: On Method: Computational Compressibility in Complex Natural and Cultural Phenomena. I’m not quite sure why I find it so pleasing. That it crosses disciplines, that’s nice: weather, microbiology, chess (AI), finance economics (asset pricing), film studies (Hollywood Economics), and cultural evolution (Macroanalysis, 19th century novel). But it’s the specific mode of the argument; it’s about description, about what kinds of things exist.

I began by framing the discussion in terms of Stephen Wolfram’s distinction between computational reducibility and computational irreducibility. I think we’ve got to look at that distinction in terms of Miriam Yevick’s 1975 distinction between holographic or Fourier logic and sequential logic. I think Wolfram’s notion of computational reducibility implies Yevick’s sequential logic. As far as I can tell, her notion of holographic logic doesn’t register with respect to Wolfram’s distinction. But it may be that what I’m calling computational compressibility (within the realm of irreducibility) resonates with her notion of holographic logic.

A random system would of course be irreducible, but that is an extreme case. The systems I looked at in that paper are not random, but the order they exhibit allows them to occupy only a relatively small region of the state space potentially open to them. Given appropriate data about the behavior of the system, that region can be identified through a computational process. Thus they are computationally compressible. The phenomenon of computational compressibility indicates order, but order of a kind that’s different from reducible order. Generative order? 

Religion & Signaling

Glenn Loury has a recent video where he distinguishes between what we might call the propositional content of an utterance and its signal value. Explains that at some length in a recent lecture he gave at Stanford, Self-Censorship, Social Information, and the Conditions of Public Reason. In the lecture he examines three cases: race in America, academic life, and Israel and Gaza. His point is that in public discourse on these topics (and others) the signal value of what one is saying often eclipses the propositional value of one’s assertions. This often results in self-censorship where a person withholds their (propositional) views for fear of signally the wrong values.

Thus, in racial discourse:

A key question in this discourse is why racial inequality persists. In this domain speech is saturated with moral meanings. Claims about inequality, education, crime, family structure, or historical responsibility are rarely heard simply as attempts to explain social outcomes. They are taken also as signals of moral orientation—of sympathy or indifference, solidarity or hostility, compassion or disdain. For many audiences, answers to that question relying on structural causes are regarded more favorably than accounts emphasizing cultural factors. Speakers adapt accordingly—many by holding their tongues.

I think lots of discourse is like that. It seems to me that much religious discourse is like this. The conservative Christian arguments against biological evolution never made much sense to me as propositional assertion, nor did the accompanying belief about the age of the earth, but as signaling, those beliefs make more sense.

I'm thinking this signaling business is about coordination, about establishing a rock-bottom level of interpersonal coordination, of common knowledge. I need to think about this some more. 

Addendum, May 2, 2026: This post about Bruno Latour's conception of modes of existence is relevant here, each with it own “felicity conditions,” as he calls them.

Silicon Valley vs. Pope Leo

I’ve only recently learned that Pope Leo XIV has decided to make AI a central concern of his papacy. Given that I think that the common Silicon Valley view of AI is idolatrous, or at best straddles the border, I’m wondering if we’re going to see a grand debate and battle between the Pope and Silicon Valley apologists. When I put the question to Claude, here’s how it responded:

Your instinct is well-founded, and Leo XIV has already moved aggressively on this front. Let me lay out what's emerged so far.

Who he is and what he's signaling

Leo XIV — born Robert Prevost in Chicago, the first American pope — identified AI as a central challenge from his very first formal address to the cardinals after his election in May 2025. His choice of the name Leo was itself a signal: Pope Leo XIII confronted the industrial revolution with Rerum Novarum (1891), defending workers' rights and the common good. By taking that name, the new pope announced that the Church would face today's technological upheaval with similar clarity.

The specific critique — and it maps closely onto your "idolatry" framing

This is where it gets interesting for your purposes. In what has been called a programmatic document on AI for World Communications Day, Leo warned that "behind this enormous invisible force that involves us all, there are only a handful of companies — those whose founders have recently been presented as the creators of the 'Person of the Year 2025.'" He expressed "serious concern regarding the oligopolistic control of algorithmic and artificial intelligence systems capable of subtly shaping behaviors and even rewriting human history."

That last phrase — rewriting human history, including the history of the Church — is remarkable. He's naming a power claim that goes beyond economic disruption.

His core argument is that "the challenge is not technological, but anthropological. Safeguarding faces and voices ultimately means safeguarding ourselves." He described face and voice as "sacred," given by God, "the constitutive element of every genuine encounter."

At a Vatican gathering that included representatives from Google, OpenAI, Anthropic, IBM, Meta, and Palantir, he said: "The person is not a system of algorithms: he or she is a creature, relationship, mystery."

The "data is not wisdom" theme

He has repeatedly insisted that "the ability to access vast amounts of data and information should not be confused with the ability to derive meaning and value from it," and that "authentic wisdom has more to do with recognizing the true meaning of life than with the availability of data." This is a direct counter to the implicit epistemology of Silicon Valley — the notion that intelligence is fundamentally information processing, that more data equals more understanding.

Where this connects to your book

Your framing of the dominant Silicon Valley attitude as idolatrous is theologically precise in a way that Leo XIV would likely recognize. The classic definition of idolatry isn't crude statue-worship — it's the attribution of ultimate value and salvific power to something that is not God. When figures like Altman, Musk, or Kurzweil speak of AI as the path to immortality, the transcendence of human limitation, the solution to death — that's structurally idolatrous in the theological sense. It substitutes a human artifact for the divine.

Leo has the resources of the entire tradition of Catholic social teaching to bring to bear on this — not just as ethics but as a rival anthropology. The Church's claim that the person is "creature, relationship, mystery" is not compatible with the claim that the person is, at bottom, a computational process that can be replicated, extended, or superseded by a sufficiently powerful machine.

Whether this becomes a genuine face-off rather than parallel talking depends on whether Silicon Valley feels the need to respond. So far, the tech executives seem to be treating Vatican engagement as a reputational opportunity — showing up in Rome, being photographed, nodding about "human dignity" — without conceding much. The more interesting question is whether Leo pushes beyond dialogue into something more like confrontation. The Rerum Novarum precedent suggests he might: that encyclical didn't merely express concern about the industrial revolution, it made specific demands.