Casino at the end of civilisation

John Oliver recently turned a segment over to platforms like Polymarket and Kalshi. The online outlet More Perfect Union has produced its own exposé. Prediction markets are moving from internet curiosity to serious political, financial and ethical debate.

People now bet on wars, elections, tariffs, deportations, recessions and ceasefires in language once kept for stocks and commodities. A missile strike, a cabinet collapse, a riot, a border clash or a failed truce can each become part of a cycle of anticipation, probability and payout. Somewhere along the way, the language of building the future quietly gave way to the language of betting on instability.

The real question is not whether these markets are moral. It is why societies are starting to find them so believable.

Part of the answer lies in a wider collapse of trust. For two decades, large numbers of people have watched confidently delivered narratives crumble against reality. From Iraq's mythical weapons of mass destruction to ever more theatrical media elsewhere, many no longer feel that official certainty deserves automatic respect.

Indian television news has shown this with force during recent tensions with Pakistan and on its domestic stories. It was once thought that only states with vast repressive machinery could keep such a grip on the narrative. Yet here is a country of 1.45 billion showing the world the way while still being called democratic and capitalist.

When propaganda outlives its half-life it produces two kinds of audience. One keeps consuming the same spectacle because identity and comfort matter more than truth. The other loses faith and turns elsewhere for scraps that feel less managed: anonymous accounts, Telegram channels, Reddit threads, leaked clips, the odd conspiracy theory.

This is one of the defining shifts of the digital age. Once people stop trusting old bodies, they rarely stop searching for certainty. They simply search elsewhere.

Prediction markets thrive in this air because they present themselves as neutral. Their pitch is simple: narratives can be staged, but money does not lie. If thousands of people risk real cash on an outcome, the market must reflect reality more honestly than a politician or an anchor.

That is a powerful pitch in low-trust societies. But studies reveal another layer beneath the talk of "wisdom of crowds". The field inside these systems is not even. Insiders, big traders, deep-pocketed liquidity holders and AI systems still hold a real edge over ordinary players. Most users end up serving less as collective wisdom and more as liquidity for better informed hands.

This also explains why these markets feel like a natural heir to the crypto era. Crypto, in its first wave, sold people a stake in future wealth. Prediction markets, by contrast, sell something darker. They make money out of uncertainty itself.

When Binance folded prediction markets into its app, the shift felt symbolic. The industry was no longer just selling belief in technological progress. It was selling wagers on disruption, conflict and instability.

This terrain turns especially volatile in countries like Pakistan and India. Huge populations, contradictory legal and economic systems, youth bulges, joblessness and permanently online political cultures create vast reservoirs of frustration and unease. Most will never have the cash to play these markets in any serious way. But they can still become the raw material of these systems.

Three parallel worlds now seem to be taking shape. The first belongs to entrenched wealth and old capital, with access to fine financial tools, private networks, AI and long-term investments. The second turns on speculative chaos: crypto swings, prediction markets, meme stocks, gambling rails and high-speed emotional trading. The third is made up of the permanently connected poor and the disaffected, who more and more serve as behavioural material for both.

For centuries, the poor mostly danced for the amusement and profit of entrenched elites. Those systems were exploitative but still held in by guardrails of institution and behaviour. Chaotic markets work differently. In systems driven by volatility, outrage and constant speculation, greed and manipulation no longer follow predictable rules. The poor become guinea pigs in a behavioural experiment on which others place their bets.

There may be another reason these markets are flourishing now. For centuries, people built their sense of worth on mastering stable domains: law, medicine, finance, engineering, writing, administration. AI is starting to outperform humans across many of these with unsettling speed. When people lose faith not only in old bodies but in the uniqueness of human expertise itself, speculation begins to look strangely tempting. If machines own the world of order, prediction and chaos start to feel like the last spaces where human instinct still matters.

Beneath all this sits a deeper shift in how societies think. Modern societies more and more prefer probability to certainty. We see this already in election forecasts, insurance models and algorithmic recommendations. Prediction markets simply push the logic further. They turn public uncertainty itself into a tradable asset.

This is where artificial intelligence enters the picture.

Much of what humans call chaos may just reflect the limits of human observation. Nassim Nicholas Taleb's Black Swan rests in part on the fact that humans miss countless variables because of cognitive limits and narrow focus. An advanced enough system may not share those limits.

An unclear image does not always mean the scene itself is unknowable. Sometimes it just means the camera is weak. Improve the lens, lift the resolution, widen the field of view and strengthen the processing, and the picture sharpens. Modern societies already give off a vast behavioural trail: cameras, transactions, internet activity, geolocation, social media. An advanced enough AI able to read those streams together may become the insider's insider.

Perfect prediction may never exist. Emergence and genuine uncertainty remain serious questions. But societies do not need perfect prediction for the balance to shift. They only need systems that look more reliable than human judgement across enough of life.

People turn to probabilistic systems because they no longer trust old bodies, governments or media. Yet the same asymmetries may return there too, this time through data scale, computing power and machine intelligence, not television anchors or political talking points.

Societies that stop believing problems can be solved often begin treating them as events to be priced instead.