Decentralized Exchange GMX Exploited for $42M, Offers Hacker 10% White Hat Bounty
A portion of the stolen funds has already been bridged from Arbitrum to Ethereum.

What to know:
- Decentralized exchange GMX was exploited for over $42M, with $9.6M already bridged to Ethereum.
- Stolen assets include $10M in Frax dollar, $9.6M in wrapped BTC, and $5M in DAI, while $32M remains on Arbitrum.
- GMX offered the hacker a 10% white-hat bounty as crypto hacks hit $2.5B in losses in the first half, per CertiK.
Decentralized perpetual exchange GMX has been exploited with over $42 million worth of crypto being stolen, according to blockchain security firm PeckShield.
So far, $9.6 million worth of the funds have been bridged over to the Ethereum blockchain, which has become a natural path for hackers who then launder funds through token mixing protocol Tornado Cash.
The rest of the $32 million remains on Abritrum, a layer-2 network that hosts the GMX exchange.
More than $10 million worth of legacy frax dollar was stolen as well as $9.6 million of wrapped btc (wBTC) and $5 million of the dai stablecoin.
GMX developers responded to the hacker, signing a message on-chain that read: "We want to offer a 10% white-hat bounty for the return of the exploited funds."
A white-hat bounty is offered to ethical hackers who find vulnerabilities in protocols and in return receive a bounty.
The exploit is another blight on a cryptocurrency industry that saw investors lose $2.5 billion to hacks and scams in the first half of 2025, according to a CertiK report.
More For You
More For You
The Genius Act ripple effect: Sui executives say institutional demand has never been higher

Evan Cheng and Stephen Mackintosh said 2025 marked a turning point for institutional adoption, with tokenization and agentic commerce emerging as the next frontier.
What to know:
- Executives cited ETF flows, DAT growth and major trading firms entering crypto.
- Tokenization and instant settlement could blur the line between traditional and decentralized markets.
- Low-latency design and composable tooling aim to power AI-driven and tokenized financial use cases.











