Nine Wallets Hold Dominant Influence Over Nearly Half of Polymarket's Dispute Votes, Bloomberg Finds
Nine anonymous crypto wallets hold dominant influence over the outcome of billions of dollars in Polymarket prediction market disputes, according to a Bloomberg analysis published May 26, 2026. The wallets, sitting among more than 6,400 participating accounts, show strongly correlated on-chain voting patterns and have sided with the winning outcome in nearly every contested case.
The finding cuts to the core of how Polymarket resolves ambiguous or contested bets. When a market closes, any party can submit a proposed outcome. If no one challenges it within two hours, the result stands. Contested proposals escalate to a vote by holders of UMA tokens, the governance token of the UMA Protocol, which Polymarket has used as its primary arbitration layer. UMA handles approximately 78% of Polymarket resolutions, with Chainlink Data Streams handling roughly 15%, making UMA's governance structure central to the platform's integrity. The system was designed to distribute truth-determination across a broad, independent base of token holders. In practice, Bloomberg found, roughly half of all effective voting power is concentrated in nine wallets whose identities remain unknown. Polymarket re-entered the United States market in July 2025 after registering with the Commodity Futures Trading Commission, making the integrity of that dispute layer a matter of direct regulatory relevance for American users.
Over the past year, nearly 2,000 Polymarket contracts were resolved through UMA's dispute mechanism. Polymarket crossed the $10 billion single-month volume threshold for the first time in March 2026, a milestone that helps explain the surge in dispute activity that followed. In April 2026 alone, approximately 230 contracts representing more than $1 billion in traded value went to a UMA vote, compared with 79 contracts in the prior six-month period. That acceleration reflects broader growth in the prediction market sector: combined monthly volume on Polymarket and Kalshi rose from under $5 billion in mid-2025 to roughly $24 billion in April 2026. The dispute layer is handling significantly more cases as volumes scale, and the concentration problem scales with it.
A separate structural concern compounds the voting power issue. More than 60% of active UMA voters over the past year have been linked to Polymarket trading accounts, meaning the same participants resolving disputes often hold financial positions in the markets they are judging. Traders have characterized this arrangement as giving anonymous whales "fact-determining power" driven by economic interests rather than neutral arbitration, a concern documented by ChainCatcher. Jan Czarnocki of Elastics put it plainly: "No serious investor will put money there as long as there's no transparency regarding the resolution criteria." That credibility deficit is reflected in UMA's market position: the governance token traded at approximately $0.45 as of May 27, carrying a market cap of around $40.7 million on a circulating supply of roughly 90.44 million tokens.
Reform efforts have moved slowly. Risk Labs, the developer behind UMA, and several ecosystem participants have initiated restructuring proposals that have largely stalled. In August 2025, UMA governance passed an upgrade called Managed Optimistic Oracle V2 (MOOV2) that created a whitelist of pre-approved addresses eligible to submit market resolution proposals, initially 37 addresses and later expanded to 177 as monthly volume crossed $10 billion. That change reduces spam submissions but does not alter who holds voting power once a dispute is elevated. The same nine wallets remain dominant at the voting stage. Polymarket said it is pursuing "ongoing improvements to transparency and market infrastructure" but has not publicly addressed the concentration of votes. Founder Shayne Coplan has acknowledged weaknesses in the system without committing to a corrective timeline.
The governance gap carries specific weight for users outside the United States. India, Brazil, Indonesia, Portugal, and Spain have all blocked Polymarket access, citing unlicensed gambling concerns. Spain's block came as recently as May 26, 2026. For the large populations in South Asia and Africa who do access the platform through workarounds, the concentration of dispute power is a concrete financial risk rather than an abstract governance debate. Kenya has explicitly blocked the platform, while Nigeria and South Africa maintain varying levels of access. In those markets, and in other parts of the continent where prediction markets are increasingly used as informal hedging tools in the absence of accessible derivatives markets, the integrity of the arbitration layer matters directly. Contested markets tied to regional events, including elections, geopolitical tensions, or local economic indicators, could be resolved by nine wallets with no stated regional knowledge or accountability. Developers in these regions building on UMA's oracle infrastructure inherit the same concentrated voting risk in their own applications.
Polymarket launched a rebuilt trading engine in April 2026, including a native stablecoin called pUSD backed one-to-one by USDC, which reduces bridge settlement risk. The company has also announced plans for a POLY token that could bring oracle functions in-house and reduce reliance on UMA entirely. As of late May 2026, no tokenomics or launch timeline for POLY have been announced. Until that path is clearer, or until the voting power distribution at UMA is meaningfully restructured, the dispute resolution system that underpins billions of dollars in monthly bets will remain concentrated in the hands of a small, unidentified cluster of wallets.