Nine Wallets Control Billions in Polymarket Disputes, and That's a Problem
A tiny group of anonymous crypto whales now decides who wins contested prediction market bets worth billions. The decentralization promise is looking pretty hollow.
画像クレジット: Image via source article. Used under fair use for news commentary. · source
Nine. That's how many anonymous cryptocurrency wallets effectively control dispute resolution on Polymarket's most contested bets, according to Bloomberg. Billions of dollars in wagers, decided by a group small enough to fit around a dinner table.
I've seen enough spec sheets to know when a system's single point of failure is hiding in plain sight. This is one of those moments.
What the numbers actually say
Polymarket has grown rapidly, becoming the go-to platform for event-driven betting on everything from elections to economic data. Kaiko CEO Ambre Soubiran noted in a Bloomberg interview that prediction markets are experiencing explosive growth, with data infrastructure becoming critical to the industry's legitimacy.
But here's the problem: when bets get disputed (and the big ones always do), resolution falls to token holders who stake capital to vote on outcomes. In theory, this is decentralized arbitration. In practice, nine wallets dominate.
We don't know who owns these wallets. We don't know if they coordinate. We don't know their methodology for deciding contested outcomes. That's an uncomfortable amount of unknowns for a system handling billions.
The regulatory mess
The SEC isn't sitting idle, though "active" might be too generous a word. The agency is delaying the launch of prediction-market ETFs that would let retail investors wager on elections and economic data through traditional brokerage accounts. SEC Chair Atkins appears to be pumping the brakes while the agency figures out how far the $15 trillion ETF wrapper can stretch.
関連記事
More in AI Models
I was asked to cover recent AI news, but what I found instead was a pile of consumer electronics listicles masquerading as tech journalism.
Aisha Patel · 1 hour ago · 4 min
Researchers are finding ways to train robots with corrective feedback and direct video imitation, potentially cutting the need for massive demonstration datasets.
James Chen · 3 hours ago · 7 min
One approach breaks expert behavior into atomic rules; the other builds a differentiable simulator from minimal real-world data. Both are trying to solve robotics' persistent generalization problem.
Aisha Patel · 3 hours ago · 6 min
A wave of new research tackles the same frustrating issue: getting robots to move smoothly when their brains can't keep up with their bodies.