A governance audit of Hyperliquid reveals foundation-run validators now control less than half of staked HYPE for the first time, yet the network's closed-source software and 27-node set leave a gap between its decentralization claims and what outsiders can verify.
A venue clearing more than $200 billion a month and holding roughly 70% of on-chain perpetuals volume is secured by 27 validators, down from a reported 81% concentration in early 2025. Foundation-run nodes now hold about 49.3% of staked HYPE, with the remaining 50.7% spread across 22 independent operators, according to the audit published July 27. The foundation operates five of the 27 validators.
"The decisive number moved this year: foundation-run validators now hold about 49.3% of staked HYPE, down from a reported 81% concentration in early 2025," the audit's author wrote. "That is a materially different network from the one described by the 81% figure still circulating in criticism."
The validator set expanded from 4 at launch to 16, then 21, 24, and 27 as of June, with registration opened to anyone and the largest stakes forming the active set. Entry has required a stake above roughly 1 million HYPE. The network generates on the order of $1 billion a year in fees, with an order book, matching engine, and liquidation system all running on those 27 machines. There is no automatic slashing anywhere in the system, leaving the unstaking queue and social consequences as the enforcement layer.
The gap that survives every correction is scale, not malice: 27 validators against roughly 1,800 on Solana and hundreds of thousands on Ethereum, securing a venue whose monthly volume exceeds $200 billion. Node software remains closed, with the foundation's promise to open the code outstanding since early 2025. Singapore's Monetary Authority added Hyperliquid to its Investor Alert List on June 26, converting the decentralization argument from a philosophical debate into a question with legal consequences.
The three powers, examined
The claim that the foundation can jail validators at will does not match the documentation, which describes jailing as peer-triggered for latency and reliability failures. The residual concern is narrower: when foundation-affiliated nodes hold close to half the stake, peer voting weighted by that stake is not fully independent. Forced upgrades are essentially accurate and largely unremarkable — every chain running a single client implementation faces the same reality. The closed-source issue stands as the most consequential: validators run a binary they cannot audit, and no distribution of stake compensates for that.
The JELLY incident in March 2025 tested governance empirically. A trader manipulated a thin memecoin market, engineering losses that landed on the protocol's liquidity vault. Validators voted to delist the market and settle it at a price favorable to the protocol, containing an eight-figure loss. The intervention worked and simultaneously answered the governance question: a market that traded on a network can be closed by a stake-weighted vote when the network's own capital is at risk, through a stake distribution the foundation then dominated.
What to watch
The stake distribution, not the validator count, is the measure that determines who actually decides outcomes. Whether foundation-run stake continues falling below 49.3% and whether any single independent operator accumulates a blocking position are the metrics that matter. The open-source commitment, outstanding since early 2025, is the single change that would most alter the audit. The next comparable intervention, and whether the decision runs through a stake distribution that no longer has a foundation majority, is the test of whether governance changed or only its arithmetic did.
This article is for informational purposes only and does not constitute investment advice.