For most of 2025, "institutional adoption of prediction markets" was a thesis. A few sophisticated traders had built positions. A few research desks were starting to cite probabilities in client notes. The Goldman piece we covered in February made clear that the analytical workflow had already shifted — Wall Street had begun pricing off prediction markets even when it wasn't acknowledging it publicly.

But the infrastructure layer underneath that shift didn't really exist. Institutions couldn't trade at size without moving prices. There was no OTC desk for prediction market exposure. No prime brokerage. No professional market making across venues. No block trade infrastructure. The probability signals were getting used; the trading mechanics were still retail.

Then last week, in roughly seven days, the institutional stack went operational.

This piece walks through what just got built, why each piece matters, and what the structural implications are for anyone trying to read prediction markets seriously — as data infrastructure, as a risk-management tool, or as a developing asset class.

The Four Pieces That Came Together

The institutional moment didn't arrive as one announcement. It arrived as four pieces of infrastructure clicking into place within a single week — each addressing a specific institutional requirement that retail prediction market platforms can't satisfy.

Galaxy Digital launched an institutional OTC prediction markets desk (June 2)

Galaxy Digital — the publicly-traded digital asset firm (NASDAQ: GLXY) — announced an institutional OTC trading desk for prediction markets through its Global Markets group. At launch, Galaxy executed a $10 million trade with crypto-native hedge fund Arca on Kalshi. The desk covers non-sports event contracts on both Kalshi and Polymarket, spanning economic, political, geopolitical, and other event-driven markets.

The structural innovation isn't the trade itself — it's everything around the trade. Galaxy clients can post USD, stablecoins, Bitcoin, or other crypto as collateral on any listed contract for positions of $100K or greater. The desk pairs prediction market positions with hedges in equities, commodities, and other assets — enabling clients to construct complete multi-asset risk strategies built around event-driven exposure. Galaxy described the launch transaction as a "first-of-its-kind crypto portfolio hedge."

For institutions, this is what was missing. A hedge fund couldn't reasonably express a $10 million view on a binary outcome through a retail-facing platform — slippage, market impact, and operational frictions made it impractical. With an OTC desk, the same view executes bilaterally at agreed pricing, with appropriate collateral frameworks and integration into existing portfolio risk systems. Galaxy is selling the picks-and-shovels of institutional prediction market participation.

Polymarket completed its first institutional block trade — on-chain (June 2)

The same day, Polymarket announced the completion of its first institutional block trade — a six-figure transaction between FalconX (a digital asset brokerage) and Anera Labs (a trading technology startup). The contract referenced the Ornn Compute Price Index, a benchmark tracking Nvidia H100 GPU rental pricing.

Two things make this trade structurally significant.

First, it settled on-chain. Polymarket's international platform operates on Polygon, and this was the first institutional prediction market trade executed on a public blockchain — the precedent for everything that follows. Block trades are privately-negotiated large transactions executed outside the public market to minimize price impact. The fact that institutional block trade infrastructure can now run on an on-chain venue, with the auditability and settlement finality that implies, is a real operational milestone.

Second, the underlying of the contract is interesting on its own. Polymarket's GPU compute price index is, in effect, a derivative on AI infrastructure costs — an asset with no public futures market and no traditional benchmark. The block trade represents an institution hedging real GPU compute exposure at scale through a prediction-market contract because no equivalent product exists in conventional financial infrastructure. The platform isn't just adding institutional features; it's becoming derivatives infrastructure for assets that lack traditional market structure.

Polymarket's block trade came roughly one month after Kalshi's first block trade — making this the second block trade by a major prediction market platform in less than 60 days, and the first executed on-chain.

Wintermute is making two-sided markets across event contracts at scale (May 2026)

While Galaxy and Polymarket grabbed the headlines, Wintermute has been quietly building the most consequential piece of institutional infrastructure: continuous two-sided liquidity provision.

The algorithmic trading firm and OTC desk announced earlier in May that it is quoting two-sided markets across event contracts on leading venues. The venues Wintermute makes markets in collectively trade more than $20 billion per month in volume as of early 2026 — a scale that didn't exist a year ago.

For anyone who has traded prediction markets at retail size, this changes the depth and quality of available execution dramatically. Institutional market makers run different books than retail traders — tighter spreads on liquid contracts, deeper books on the major events, and crucially, professional risk management that keeps quoting through volatility instead of pulling away when markets get interesting. Wintermute's participation across platforms is what makes serious institutional execution possible. It is the closest analog prediction markets have to designated market makers in equities or primary dealers in fixed income.

Kalshi cleared $17 billion in May — and is openly pivoting to Wall Street

Kalshi processed more than $17 billion across event contracts in May 2026 — a 2,500% year-over-year increase. The growth came from retail momentum that had been building since September 2025, but the strategic direction has now clearly pivoted to institutional adoption.

The institutional moves are concrete:

• A partnership with Fidelity National Information Services (FIS), the global financial-technology infrastructure provider, to develop the connectivity that institutional trading systems require

• A partnership with Ark Invest announced in March 2026 to develop specialized prediction markets explicitly for institutional investment analysis

• A dedicated institutional desk with Andy Ross leading institutional sales (formerly at Morgan Stanley)

• The first block trade infrastructure on a CFTC-regulated platform, predating Polymarket's by roughly one month

CEO Tarek Mansour put the institutional case directly in CNBC's coverage: "Event contracts could become a trillion-dollar market." That's not a marketing line — it's the trajectory the operational metrics support if institutional volume continues building on the retail base.

What's driving the institutional interest, per Kalshi's head of institutional Andy Ross, is hedging. Instead of trying to game financial-market reactions to alleviate risk around specific events — an election, an economic data release, a regulatory decision — a firm can directly trade a binary contract on the event itself. "Those are tradable assets now that people can directly trade upon," Ross told CNBC, "as opposed to trading on a derivative of those."

That's the institutional thesis stated cleanly. Event contracts are tradable assets. Treat them accordingly.

What Just Got Built, Structurally

Step back from the individual announcements and the picture is unmistakable. The four pieces are not independent — they are the components of a complete institutional trading stack:

This is what a functioning institutional asset class looks like. Equities had specialists, broker-dealers, market makers, OTC desks, and block trading infrastructure for decades before retail trading scaled. Fixed income had primary dealers and OTC voice trading. Crypto built equivalents painfully over a decade. Prediction markets just compressed an equivalent buildout into roughly one quarter.

Six months ago, asking "is there an OTC desk for prediction market exposure?" would have produced a confused look. The answer now is "yes, multiple, with documented multi-million-dollar prints and integrated multi-asset collateral frameworks." That sentence is the institutional moment, stated plainly.

The Goldman Thesis Just Got Operational Confirmation

In February we wrote about Goldman, Wolfe, and Jefferies starting to publish research that tracked prediction market probabilities almost line for line. The point of that piece was that Wall Street had quietly begun pricing off prediction markets even when the practice wasn't being acknowledged publicly.

What just changed isn't that thesis — it's the operational layer underneath it. In February, the institutional engagement with prediction markets was analytical. Reading the markets. Citing the probabilities. Using the signals as inputs to traditional research workflows. The trading itself, where it happened, was either retail-sized or routed through bespoke arrangements.

Now the analytical engagement and the trading engagement are merging. The same institutions that were reading prediction markets six months ago can now execute meaningfully sized positions through Galaxy's OTC desk. They can hedge real GPU compute exposure through Polymarket block trades. They can rely on Wintermute's two-sided quotes for working orders. They can integrate Kalshi's regulated platform into existing trading systems via the FIS partnership.

The retail-to-institutional transition that took crypto a decade has taken prediction markets less than a year. The implications cascade quickly: as institutional volume grows, retail signal quality improves (because professional capital arbitrages obvious mispricings), which makes the markets more useful as analytical inputs, which pulls in more institutional engagement. The flywheel is now visible and running.

What This Means for Anyone Reading Prediction Markets Seriously

Three implications worth sitting with.

First, the data quality question changes. When prediction markets were primarily retail venues with thin liquidity on most contracts, much of the "noise" in cross-venue divergences came from inefficient pricing — retail traders on Platform A simply hadn't seen what retail traders on Platform B had figured out. With institutional market makers like Wintermute providing continuous quotes, that source of inefficiency starts to disappear quickly. The remaining cross-venue divergences become more structurally interesting: they reflect real differences in resolution criteria, real differences in trader populations' information, or real arbitrage opportunities that survive institutional attention. The noise floor drops, which makes the signal more readable. (And makes it more important to understand which differences are structural and which are arbitrageable — which is exactly the analytical layer this infrastructure now requires.)

Second, the regulatory landscape gets sharper. Institutions don't deploy meaningful capital into venues with ambiguous regulatory status. Galaxy's OTC desk operating across Kalshi (CFTC-regulated, US) and Polymarket international (not US-available) implies the firm has done the regulatory work to support each side cleanly. Wintermute making markets across venues implies the same. As more institutions enter, the regulatory frameworks around event contracts harden — the seventeen-state challenges, the CFTC jurisdiction questions, the gambling-vs-derivatives distinctions — because real money requires clarity. The Congress investigation we covered in May becomes one input into that hardening, not its endpoint.

Third, the analytical infrastructure underneath the trading infrastructure is now the bottleneck. Galaxy's OTC desk needs to quote spreads. To quote spreads, you need cross-venue reference pricing. To produce reliable cross-venue reference pricing, you need normalized data from all the platforms an institution might want exposure to, with resolution metadata, liquidity quality scoring, and structural reconciliation across venues. That data layer — the institutional Bloomberg-equivalent for prediction markets — doesn't ship with the trading infrastructure that just got built. It has to be built independently, and it has to be built by someone whose interests are aligned with all institutional participants rather than with any single platform.

That last piece is the part we're paying close attention to, because we're building it.

The Bigger Picture

The "prediction markets are growing up" narrative has been around for years. Every six months a wave of coverage suggests institutional adoption is around the corner. Most of those waves have been premature.

This one isn't. The evidence is no longer rhetorical. It's operational: documented OTC trades, documented block trades, documented market making at scale, documented volume figures, documented prime brokerage relationships, documented regulatory partnerships. The infrastructure isn't being announced. It's being used.

For institutions, the implication is straightforward: prediction markets are now a tradable asset class with the operational infrastructure to support institutional position-taking. For platforms, the implication is that the next phase of competition is about institutional features — execution quality, regulatory clarity, integration partnerships, market making depth, settlement reliability — not retail UX. For everyone watching from the analytical seat, the implication is that the signal quality of these markets is about to improve materially, while the analytical layer needed to make sense of cross-venue divergence becomes more valuable, not less.

A year ago, the question was whether prediction markets would matter. That question is now answered. The new question is what gets built next on top of operational institutional infrastructure — and what data layer the institutions now active in this market will rely on to understand it.

This is the fifteenth installment in the Assymetrix Intelligence Brief series.

Previous: "Polymarket Is Right About Footprints. The Question Is Who Else Can Read Them."

Related: "Goldman Said the Quiet Part Out Loud: Wall Street Is Now Pricing Off Prediction Markets" — the predecessor on the institutional engagement thesis.

Assymetrix is building the cross-venue, on-chain intelligence layer that turns public ledgers into readable, structured market data — independent of any single platform.

assymetrix.com/blog

Note on sources: Galaxy Digital's OTC desk launch and the $10M Arca trade are from Galaxy's June 2, 2026 press release and Markets Media coverage. Polymarket's first on-chain block trade is from CNBC's June 2, 2026 reporting, attributed to Brooke Rizzetto, head of institutional liquidity at Polymarket. Kalshi's $17B May 2026 volume and the 2,500% YoY figure are from CNBC's June 1, 2026 coverage. The Wintermute multi-platform market making is from the Markets Media reporting on institutional infrastructure growth. The Kalshi-Ark Invest partnership detail is from March 26, 2026 announcements. Andy Ross is Kalshi's head of institutional. Tarek Mansour's "trillion-dollar market" quote is from CNBC coverage. All figures verified against primary sources where available.