Hemi (@hemi_xyz): The Resolv USR Incident: A Cautionary Tale – And Why Disciplined Derisking Is Everything

The Resolv USR Incident: A Cautionary Tale – And Why Disciplined Derisking Is Everything On March 22, 2026, Resolv’s USR stablecoin suffered a major incident. A single compromised offchain privileged key allowed an attacker to mint approximately 80 million unbacked USR tokens using only around $200,000 in collateral. The attacker rapidly dumped the tokens on DEXs, extracting roughly $23–25 million in real crypto and driving the price down to as low as $0.02 in some pools. The underlying collateral pool remained intact. Resolv’s statement: https://x.com/ResolvLabs/status/2035830314799599616?s=20 Unlike many past incidents, this wasn’t driven by a classic smart contract vulnerability like reentrancy or oracle manipulation. Rather, it stemmed from more fundamental design choices, particularly a single key controlling critical minting rights and autonomous systems without proper safeguards. The initial mint triggered contagion into several Morpho USDC vaults holding USR collateral. The damage was significantly amplified by Morpho’s Public Allocator, an autonomous entity that continued allocating fresh capital into failing markets instead of stepping back. Omer Goldberg’s thread offers one of the best postmortems on how the allocator behavior worsened the outcome: https://x.com/omeragoldberg/status/2035817791786221990 In response to the incident, Resolv acted quickly to pause operations and begin its investigation. In their statement, it was noted that the collateral remained safe and the issue was isolated to the issuance. Several protocols avoided exposure entirely. Aave, for instance, had zero direct impact thanks to strong risk parameters, including Hemi partner @OdysseyFi. https://x.com/DegenerateNews/status/2035634287307866230 Shoutout to Odyssey CEO @jeffthedunker and @manojpatidarr in Engineering for their excellent upfront development, thorough due diligence on yield sources, proactive monitoring, and smart, fast reaction. They kept any negative impact minimal to zero. True autonomy doesn’t mean blindly turning over critical operations to failure-prone processes. It means designing infrastructure and intelligence to work in lockstep so efficiency never becomes an insecure shortcut. Why This Matters for How We Build at Hemi Events like this reinforce why we take such a deliberate and disciplined approach to every aspect of our engineering process at Hemi. We don’t design systems where a single EOA can mint unlimited tokens, and we never will. That kind of single point of failure is exactly the risk we actively design against from day one. This is exactly why we design hard: rigorous audits, extended Testnet testing, private testing rounds after that, and then a careful, staged ramp-up of capital and liquidity. We start small, closely monitor performance, prove stability at every layer under real conditions, and only then gradually add more. Every one of these steps represents intentional, real-world derisking. Turning theoretical edge cases into proven resilience long before meaningful capital is at stake. We saw this same disciplined mindset in action with the Odyssey team. Ultimately, no one wants incidents like this to happen, but how you design ahead of time and how you react when pressure hits can make a world of difference.