Judge Acquits MIT Brothers in $25 Million MEV Case, Leaving DeFi Law Unsettled
Key Takeaways
- •The jury determined that prosecutors failed to prove the material misrepresentation element required to establish wire fraud charges against the Peraire-Bueno brothers.
- •The brothers used software called Omakase to attract sandwich bots with bait trades and exploited an MEV-Boost relay flaw to access private transaction bundles in a 12-second operation prosecutors valued at $25 million.
- •Flashbots' post-mortem analysis estimated the exploit's losses at roughly $20 million, blamed a relay that released private transactions after receiving a bad block header, and reported the flaw has been fixed.
- •Judge Clarke distinguished Ethereum from regulated exchanges, noting that written trading rules can impose obligations on traders, but prosecutors failed to demonstrate such obligations applied, so other MEV cases may still be resolved differently.
- •MEV-related legal uncertainty continues to weigh on institutional products such as BlackRock's iShares Staked Ethereum Trust ETF, which will stake Ether only when it deems legal and regulatory risks acceptable, while an AFT study tracked $233.8 million extracted from 7.2 million CEX-DEX arbitrage trades.

On November 16, US District Judge Jessica Clarke found MIT-educated brothers Anton and James Peraire-Bueno not guilty in a case tied to an alleged $25 million Ethereum exploit. According to her ruling, the jury concluded that prosecutors had not proven the material misrepresentation required to establish wire fraud claims. Although the verdict ends the matter, it does not clarify the legality of other MEV strategies. MEV, or maximal extractable value, is profit gained by inserting, excluding, or reordering transactions within a blockchain block.
Why the deception charge collapsed
In 2025, a jury failed to reach a verdict after four weeks of testimony, leaving an acquittal motion pending before Clarke. Her inquiry then turned to the brothers' bait orders. The disputed transactions had been completed rather than withdrawn, and they were executed without any guarantee that other trades would be grouped together or follow a particular order. Prosecutors could not identify any trading rules that required such guarantees.
Clarke stressed that the acquittal does not amount to an endorsement of the brothers' trading methods. "The question before the court is not whether the Peraire-Buenos' conduct was right or moral, or even whether it violated uncharged crimes…" she wrote.
How Omakase drained the sandwich bots
The brothers used a program called Omakase to lure sandwich bots with bait trades. A sandwich bot is an automated trader that places orders before and after another user's pending trade to profit from the resulting price swing. Once the bots responded, the pair exploited a flaw in an MEV-Boost relay — the middleware that carries private transaction bundles to validators — to view private transaction bundles and replace the bots' trades. Prosecutors said the 12-second operation yielded $25 million.
Flashbots' post-mortem analysis estimated the losses from the exploit at about $20 million, attributing the incident to an MEV-Boost relay that released private transactions even after receiving a bad block header. The flaw has since been fixed. The $20 million figure covers the same exploit described by prosecutors and is not an additional loss.
What the acquittal does not settle
The verdict does not imply that every MEV strategy is now legal. In her ruling, Clarke drew a distinction between Ethereum and regulated exchanges, where written rules can forbid deceptive orders. Such rules can impose obligations on traders — obligations that prosecutors failed to demonstrate in the Peraire-Bueno case. Other MEV cases may be resolved differently, particularly if traders acted under clear exchange rules or violated other laws.
DeFi advocates had feared the prosecution could turn unwritten expectations about validator behavior into criminal duties. Coin Center argued that Ethereum's technical checks establish whether a validator followed protocol rules, not whether that validator treated other traders fairly. As Cryptopolitan reported, the DeFi Education Fund also sought to file a brief, warning that uncertain criminal liability could discourage developers and push DeFi activity overseas. Prosecutors opposed the filing, saying it added nothing relevant to the pending acquittal motion. Both groups wanted clearer legal boundaries so developers and validators could understand their risk of criminal prosecution.
Why institutional staking is exposed
Uncertainty over MEV rules also matters to institutions that stake Ether. BlackRock's SEC-filed prospectus says its iShares Staked Ethereum Trust ETF uses staking service providers and will stake Ether only when the sponsor considers the legal and regulatory risks acceptable. The filing also warns of slashing, an Ethereum penalty that can cost validators their staked Ether for violating consensus rules, such as submitting conflicting attestations. Slashing is separate from the alleged MEV-related fraud examined in the brothers' case. Institutional funds therefore face two distinct concerns: potential losses from validator penalties, and uncertainty over the laws governing trading activity.
MEV's cost to traders, and who controls the blocks
ESMA warned in 2025 that MEV can move value from users to extractors and undermine market fairness, and that existing safeguards remain inadequate. A separate AFT study tracked $233.8 million extracted from 7.2 million CEX-DEX arbitrage trades by 19 searchers, with three participants capturing about 75% of that value.
That extraction flows through a small set of infrastructure: searchers spot the opportunities, specialized builders assemble them into blocks, and MEV-Boost relays pass those blocks to validators for publication — the same relay pipeline whose flaw figured in the Peraire-Bueno case.
For global markets, the question is whether clearer trading rules and stronger block infrastructure can curb those costs without shutting out decentralized competition.