The Privatization of Risk

On May 19, 2026, two announcements landed within hours of each other that, taken together, describe something larger than either story in isolation. The first: President Trump called on AI companies to build, bring, or buy 100% of the energy needed for their data centers as part of what his administration is calling the "Ratepayer Protection Pledge." The second: Polymarket, the on-chain prediction market platform, launched markets tied to private company outcomes using Nasdaq Private Market data. These items ran under separate headlines and reached different audiences. Both, however, represent the same underlying logic: move risk off public books and onto private actors, then call it a policy success.

The Ratepayer Question

The framing of Trump's announcement is straightforward enough: ordinary electricity customers should not subsidize the power demands of an industry that is, by any measure, extracting enormous value from the grid. Data centers for artificial intelligence companies consume staggering amounts of electricity — enough that utilities in Virginia, Texas, and Georgia have already begun renegotiating industrial rate contracts under pressure from residential customers and local governments. The Ratepayer Protection Pledge, in this reading, is a consumer-protection measure. It asks the private sector to internalize a cost that would otherwise fall on households and small businesses.

But the pledge's name obscures what it actually does. By demanding that AI companies cover 100% of their energy needs, the administration is not eliminating a subsidy — it is renegotiating who bears the risk of building out grid infrastructure to meet accelerating demand. Power generation and transmission capacity do not materialize because a corporation writes a check. Utilities, state regulators, and federal siting authorities still make the decisions about where power comes from and how it moves. The pledge transfers the financial exposure of new capacity onto AI firms while leaving the institutional decisions about that capacity with the same public bodies that have struggled to permit new generation quickly enough. The result may well be that AI companies fund the connections to the grid without funding the grid itself — a distinction that matters enormously when power shortages occur.

There is a second layer to consider. The companies capable of meeting a 100% energy requirement are not startups. They are Microsoft, Google, Amazon, and Meta — firms with the balance sheets to self-build or long-term-contract generation capacity. The pledge, in practice, draws a larger moat around incumbents and makes it harder for smaller players to compete. That is not a bug in the policy; it may be the point. But it deserves to be named as a structural effect rather than buried under the language of consumer protection.

When Prediction Markets Meet Private Companies

The Polymarket announcement on the same day received less mainstream attention but raises more immediate questions about market integrity. Polymarket has launched prediction markets tied to private company outcomes, sourced from data provided by Nasdaq Private Market. For the uninitiated: prediction markets allow participants to trade on the likelihood of specific future events — in this case, whether a private company will achieve certain milestones, file for an IPO at a given valuation, or complete a merger. The Nasdaq Private Market data gives Polymarket access to pricing signals from secondary markets in shares that have never been publicly listed.

The novelty here is not the existence of prediction markets — those have operated in various forms for decades — but the convergence of on-chain trading infrastructure with private-company data streams. It creates, for the first time at scale, a market mechanism for aggregating dispersed information about companies whose financials are not publicly disclosed. That is not inherently sinister. Markets are, at their core, information-processing machines. The question is what information, processed by whom, and to whose benefit.

Private company prediction markets introduce several asymmetries that their operators have not fully addressed. Institutional investors and insiders who understand a company's true trajectory can take positions in these markets before that information is reflected in secondary share prices. Retail participants, who make up the bulk of Polymarket's user base, are operating with a thinner informational edge. The Nasdaq data is real-time enough to matter but lagged enough that sophisticated actors can front-run the signal. This is not fraud — it is legal under current regulatory frameworks — but it is a structural redistribution of advantage toward those who already know more.

The Risk-Transfer Economy

What connects these two stories is not their subject matter but their underlying theory of governance. Both assume that private actors, when given clear incentives and sufficient latitude, will make better decisions than public institutions. Trump's energy policy shifts risk onto corporations in exchange for deregulatory flexibility. Polymarket's expansion monetizes information asymmetries in the name of market efficiency. Neither premise is obviously wrong in isolation. Together, they describe an economy in which the upside accrues to private capital and the downside is quietly socialized — through grid reliability obligations, through bailouts when large AI firms overextend, through market corrections when prediction markets on private companies produce outsized payouts for early movers.

The pattern is not new. It has defined financialized capitalism for decades. What has changed in 2026 is the scale and the speed. AI infrastructure commitments are measured in gigawatts and decades-long power purchase agreements. Prediction markets on private companies are measured in millions of individual trades executed on-chain, largely beyond the reach of existing securities law. The regulatory frameworks that might govern these activities — utility regulation, securities regulation, antitrust oversight — were designed for slower-moving, less capital-intensive industries. They are catching up, but they are not caught up.

Who Owns the Downside

The honest answer to that question is that we do not yet know. The Ratepayer Protection Pledge is days old; its effects on utility planning, data center siting, and AI company investment decisions will take years to materialize. Polymarket's private-company markets launched on May 19, 2026 — their trading volumes, participant demographics, and informational properties are not yet established data points. The articles Monexus reviewed do not contain enough information to assess either initiative's likely outcomes with confidence.

What can be said with reasonable confidence is that both announcements reflect a deliberate choice about where to place risk in an economy that is simultaneously demanding more electricity, more data infrastructure, and more market-based mechanisms for pricing uncertainty. That choice — to privatize risk while leaving public institutions responsible for managing the consequences when things go wrong — is a political decision dressed in the language of efficiency. It deserves scrutiny that goes beyond the press release.

The ratepayer is told they are being protected. The prediction market participant is told they are participating in market efficiency. Both claims deserve to be tested against the structural incentives that these policies create, not the framing that accompanies them.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://t.me/Cointelegraph/15234
• https://t.me/Cointelegraph/15235