Most prediction markets rely on external oracles to tell the blockchain what actually happened in the real world. Hyperliquid just decided to fire that referee and let its own validators call the shots.
The platform launched validator-governed outcome markets on May 25 under an expanded version of its HIP-4 framework. The system allows Hyperliquid’s 24 validators to publish, deploy, and settle prediction-style contracts entirely through on-chain votes, with no external oracle required.
How it works, and why it matters
Polymarket, the biggest name in crypto prediction markets, relies on UMA’s oracle system to resolve bets. That introduces a dependency. If the oracle misbehaves, gets manipulated, or simply goes offline at a bad time, the entire market resolution process breaks down.
Hyperliquid’s approach collapses that external dependency into the same set of validators already securing the network. These 24 validators sign blocks roughly every 70 milliseconds while managing over $3 billion in deposits. Now they also vote on whether real-world events have occurred.
The validators use automated newsfeed software to guide their voting on market deployment and outcomes. Developer Yaigourth announced the change, stating that the validator set now serves as the oracle for prediction markets.
The contract design
The HIP-4 outcome contracts are fully collateralized binary instruments that settle to either 0 or 1. No leverage. No liquidation risk. You buy exposure to an event happening or not happening, and the contract resolves cleanly when validators reach consensus on the outcome.
The contracts enable exposure to events like economic data releases, and these zero-leverage contracts settle inside the platform’s existing trading infrastructure.
Competitive landscape and token performance
Polymarket built its prediction market on top of Polygon with UMA handling dispute resolution. Kalshi operates as a regulated exchange with centralized settlement. Hyperliquid is attempting something different: a closed-loop system where the same validator set that secures the chain also resolves markets, with no external dependencies at any layer.
The HYPE token was trading near $62 on the day of the HIP-4 update. That price reflects a gain of more than 36% over the prior week and over 50% in the preceding 30 days. Ongoing fee buybacks have totaled over $1.16 billion cumulatively, creating persistent buy pressure on the token.
When a high-stakes prediction market has a disputed outcome, 24 validators voting on resolution is a very different dynamic than a specialized oracle network with its own economic incentives for honest reporting. The system works cleanly when outcomes are obvious. The interesting test comes when they’re not.