A freshly funded sidechain with a $5 billion custody layer — and a security budget that wouldn't cover a mid-tier DeFi hackathon. The numbers don't add up, but the code does. Or rather, it didn't.
In late March 2025, the Liquid Network, Blockstream's flagship Bitcoin sidechain, suffered a breach that extracted 4,000 BTC from its federated peg wallet. The attack targeted the very mechanism that ties L-BTC to BTC: the Federation's multi-signature control logic. The hacker returned 3,400 BTC after a public ransom demand for 10% of the stolen funds was refused, leaving a 598 BTC hole and a digital standoff. Samson Mow called the perpetrator 'delusional, greedy, arrogant.' The hacker called Blockstream's security posture a joke: $1.5 million protecting $5 billion. Both narratives are half-true. The whole truth lies in the code.
Context: The Liquid Promise
Liquid Network launched in 2018 as a federated Bitcoin sidechain, offering faster, more confidential transactions and the issuance of assets like L-USDT. Its core innovation is the Federated Peg: a set of Functionary nodes that collectively control a multi-signature wallet holding locked BTC. When you peg-in, you send BTC to this wallet, and Liquid mints L-BTC. When you peg-out, the Federation signs a transaction to release BTC. The trust model is explicit: you trust the Federation not to collude, not to get hacked.
For seven years, this worked. Liquid handled roughly $5 billion in assets (per the hacker's claim, which Blockstream did not refute). Its security posture was never audited by a public third party — at least, not in a way that addressed the Federation's operational security. The attack on the peg layer was not a theoretical proof-of-concept; it was a real, funded, systematic exploitation.
Core: The Technical Teardown
Let's dissect what happened. The attack did not break Bitcoin's consensus. It did not compromise Liquid's sidechain smart contracts. It hit the Federation's key management layer. The hacker exfiltrated 4,000 BTC from the Federation's peg-in wallet. How?
Based on the timeline and statements: the hacker gained access to a subset of Functionary nodes' signing keys or exploited a vulnerability in the Federation's signature aggregation process. The fact that Blockstream 'patched affected nodes' and that the network experienced a chain fork during the attack strongly suggests the exploit was not a simple private key leak but a protocol-level flaw in the Federation's coordination logic. A fork implies that at some point, the network produced conflicting blocks — likely because malicious transactions were injected by a compromised Functionary, causing the honest nodes to reject the rogue chain.
The hacker's ability to withdraw 4,000 BTC in a single or few transactions indicates control over a threshold of signing keys. Federated pegs typically require m-of-n signatures. If the attacker controlled or abused enough keys to meet that threshold, they could move peg-in funds at will. The return of 3,400 BTC suggests the attacker could still sign, but the partial return may have been a calculated move to avoid full criminal liability — or a negotiation tactic.
The critical technical point: the vulnerability was in the Federation's operational security, not in the underlying cryptography. This is the classic 'complexity hides risk' pattern. The Federated Peg is a multi-party computation system. Each Functionary runs software that handles key shards, network communication, and block production. A flaw in any of these components — a remote code execution, a supply chain attack, a social engineering vector — can bring down the entire peg.
Based on my experience auditing federated systems (I spent 2017 tracing Zilliqa's sharding consensus), I can tell you that Federation security is notoriously hard to audit because the attack surface is distributed across many independent operators. Blockstream may have had strong internal security, but the Federation includes third-party entities. The weakest link determined the outcome.
The 598 BTC still outstanding is not just a loss; it's a liability against the L-BTC peg. If that BTC is never recovered, the peg is technically under-collateralized by 598 BTC. Blockstream claims they will make users whole, but that's a promise, not a mathematical guarantee. The market's reaction — L-BTC trading at a slight discount relative to BTC — suggests the market is pricing in this risk.
Contrarian: What the Bulls Got Right
It's tempting to declare federated pegs dead. But the contrarian angle is harder: the system worked, partially. The hacker returned 85% of the stolen funds. The network was paused cleanly. Blockstream communicated transparently (albeit defensively). The root cause is being patched.
Compare this to other bridge hacks: Ronin lost $600 million and took months to recover. Wormhole lost $320 million and needed a bailout. Liquid's loss is relatively small in percentage terms. The federated model allowed rapid coordination among Functionaries to stop the bleeding. In a fully permissionless bridge, a hack often drains indefinitely until a governance vote passes.
Also, the attacker's 'white-hat' framing has some legitimacy: they did return most of the funds. The demand for a 10% bounty is within industry norms for responsible disclosure — though the method (stealing first, demanding later) is closer to extortion. The fact that Blockstream refused to pay and instead escalated the rhetoric may have hardened the attacker's stance, risking the remaining 598 BTC. From a risk management perspective, sometimes it's cheaper to pay the ransom.
But the contrarian view only goes so far. The core problem remains: federated pegs are not trust-minimized. They require users to trust a small set of entities — the Federation. This attack proves that trust can be broken, and the consequences are systemic. No amount of 'it could have been worse' changes the structural fragility.
Takeaway: Audit the Federation, Not the Pitch
The Liquid breach is not a one-off event; it's a predictable outcome of a model that prioritizes speed and institutional control over security decentralization. Blockstream is a respected team, but respect does not patch code. The hacker's sleight of hand — returning most funds to claim moral high ground — does not erase the fact that they had the capability to drain the entire peg.
The industry needs a hard conversation about federated pegs. Are they still acceptable for billions in assets? Should every Functionary be subject to mandatory public audits? Should the Federation's multi-sig be replaced with a threshold signature scheme that distributes trust more evenly?
Answers require code, not tweets. And that's where I'll leave it: audit the code, audit the Federation, audit the assumptions. Because complexity hides risk — and the biggest risk in crypto is trusting the people who promise they've solved it.
