The Gambling Hall

It is agreed that studying [the laws of games of chance] contributed to the discovery of the theory of probability, to topology and the theory of strategic games. But they are not regarded as capable of providing a model for depicting the real world or unwittingly structuring a kind of embryonic, encyclopedic knowledge of it. Moreover, fatalism and strict determinism, to the degree that they deny free will and responsibility, view the entire universe as a gigantic, general, obligatory, and endless lottery in which each drawing inevitably implies the possibility or even the necessity of participating in the next drawing, and the next, and the next, ad infinitum.

Also, among leisure classes whose work is insufficient to absorb their energies or occupy all their available time, games of chance frequently acquire an unexpected cultural significance which influences their art, ethics, economy, and even life experience.

— Roger Caillois, Man, Play and Games (1958)

Gambling converts time into a narcotic.

— Walter Benjamin, Paris, Capital of the 19th Century (1938)

In media theorist Sun-ha Hong’s research, contemporary shipping and industrial workers have their discretion extracted by predictive systems and reach a level of alienation from their work that looks something like the working underclass of Metropolis.¹ With every move surveilled and predicted, they lose all agency and become Maschinenmensch:

[…] automation separates the worker from their skills and knowledge - a separation which then enables the factory to operate with deskilled workers, and to turn that information around to evaluate and control worker behaviour. In Amazon warehouses, human ‘stowers’ handle incoming items of staggering variety, grouping them into bins in giant ‘pick towers,’ a task that involves a certain degree of discretionary judgment. When this process is laced with data-gathering systems, however, each worker action - such as the stower’s scanning of the item, their rate of pieces per hour, and their bathroom breaks – provides data grist for improving predictions of the stowing process. Individual judgments and tacit knowledge are siphoned into a unique source of knowledge for managers, but not for the workers themselves. The key innovation is not merely to pack seven boxes instead of six, but ensuring that it is the manager who can set the ‘rate’ of seven, or eight, or two hundred, in ways that are precise for the manager, and opaque to the worker. (Hong 2023)

The current de-skilling of information workers through the mass adoption of machine-learning coding tools also feels in some ways like an echo of the industrial revolution, in which craftspeople were marginalized and siloed to roles within a larger process of production over which they had less insight or discretion than before. My experience of using machine-learning coding tools feels very much like pulling the arm of a slot machine that can either spit up fully-formed functioning code that solves the problem of the moment or can stuff a project full of the software equivalent of asbestos—a future problem to be untangled, hidden by a veneer of “what you asked for” functionality. In either case one pays rent for the convenience of offloading cognitive labor and misses out on the learning process of navigating that difficulty. When I see interventions from the tool that happen within the realm of my own expertise, I tend to be repulsed and I reject them. When the tool writes code outside of my skillset, I assume that if it works, it must be good-enough. This leads to a frictionless state where I never get “stuck” on something I don’t know how to do. Pulling the slot machine arm can feel addictive—someone without any coding experience abandons their discretion entirely to the imposing speed and apparent confidence of the tool.

Paul Lafargue, whom Benjamin quotes at length in The Arcades Project, saw the modern bourgeois as a professional gambler:

Modern economic development as a whole tends more and more to transform capitalist society into a giant international gambling house, where the bourgeois wins and loses capital in consequence of events which remain unknown to him. . . . The ‘inexplicable’ is enthroned in bourgeois society as in a gambling hall. . . . Successes and failures, thus arising from causes that are unanticipated, generally unintelligible, and seemingly dependent on chance, predispose the bourgeois to the gambler’s frame of mind. . . . The capitalist whose fortune is tied up in stocks and bonds, which are subject to variations in market value and yield for which he does not understand the causes, is a professional gambler. The gambler, however, . . . is a supremely superstitious being. The habitués of gambling casinos always possess magic formulas to conjure the Fates. (Benjamin 2002, 276)

Prediction markets are online exchanges where participants wager on the outcomes of future events—elections, wars, economic indicators, celebrity gossip—buying and selling contracts that fluctuate like stocks. The CEO of Trump-backed prediction market Kalshi has described his company’s mission as “replacing debate, subjectivity, and talk with markets, accuracy, and truth” (“Kalshi Reaches $11 Billion Valuation as App Takes over America” 2025). This is the next intensification of what Benjamin diagnosed as the triumph of “information” over other narrative forms: where information arrived pre-explained and always plausible, the prediction market compresses it further—into a single number, a probability, stripped of context and offered as self-evident truth. Hong calls prediction a relational grammar: a set of epistemic expectations governing how facts are made, what counts as knowledge, and whose way of seeing is imposed in the process (Hong 2023). The very act of declaring something predictable conceals ambiguity while appearing to claim nothing at all. In the past two years Kalshi’s monthly trading volume has exploded from roughly one billion dollars to over twenty billion, pressing into domains once considered too morbid to commodify.² On Kalshi-competitor Polymarket, war markets allowed apparent insiders to profit from advance knowledge of military strikes on Venezuela and Iran; bettors who stood to lose money on the outcome of an Israeli interception threatened the life of a journalist whose reporting contradicted their wagers. The idea is not new to the military—in 2001, DARPA funded FutureMAP, a proposed market where traders would bet on coups, terrorist attacks, and assassinations in the Middle East, but it was killed by congressional outrage within forty-eight hours of its public disclosure (Futrelle 2026). What presents itself as collective intelligence is the gambling hall dressed in the language of epistemology—a mechanism for concentrating the power to define the situation in the hands of the predictor, at the expense of those whose lives are being predicted upon.

Both the gambling managerial class and the Maschinenmensch under their control are operating in some altered temporality of continually-interrupted memorylessness. The “future” in this mode is not a horizon of possibility but a perpetual wager under the heavy shadow of a resurrected, primordial fate.

The idea that time is a perpetual motion machine for generating ‘value,’ that boom and bust cycles of uncertainty are a kind of casino for generating infinite wealth, is “gamerthink”—an intrusion of the symbolic world of tokens onto the material world it claims to represent, to the detriment of the material world. It is the epistemological stance that all systems are fundamentally rule-governed, that optimization is always possible, and that the right strategy (or exploit) will eventually master any domain. Applied to finance or geopolitics, it produces the same superstitious fatalism Lafargue describes.

Benjamin, writing in 1936, saw clearly where the logic of technological overproduction leads when decoupled from the reorganization of property relations:

If the natural utilization of productive forces is impeded by the property system, the increase in technical devices, in speed, and in the sources of energy will press for an unnatural utilization, and this is found in war. The destructive power of war furnishes proof that society has not been mature enough to incorporate technology as its organ, that technology has not been sufficiently developed to cope with the elemnental forces of society. (Benjamin, Bullock, and Jennings 2004)

My point here is not to moralize about individual technologies, which doubtlessly have potential benefits under a different regime and with cost transparency. One major problem is the repeating cycle in which collections of disparate technologies are bundled together and given an enchanted name like a colony of rats in a trenchcoat and hat—”Cybernetics,” “Expert Systems,” “Blockchain,” “Web3,” “Artificial Intelligence”—to furnish a renewed idol into which the gambling owner class can pour resources.³ Each new magic box promises to spin straw into gold and to solve the myriad growing problems caused by its own production. When it fails, the idol is rebranded or rebuilt, and if no rebrand will suffice—if the bubble has inflated beyond what civilian markets can absorb—the war industry offers its final form.⁴

The tower is always being built and the tower is always falling. The only future that capitalism and its various subordinate religions and nationalisms can imagine is eternal growth toward a collapse that should never, but must someday, occur. This is the structure of the gambling hall: the game must continue; the house always eventually wins; and the gambler, unable to leave the table, mistakes their compulsion for a challenge to fate. The name of the magic charm is continually changing in a cycle of fashion and novelty, a technological enchantment that refinances the planet periodically, dedicated to cancerous growth rather than addressing the needs of humanity and the world it stewards.