Points of Focus
- Attackers drained nearly 4,000 BTC without compromising Bitcoin itself.
- Unbacked L-BTC exposed a critical weakness in sidechain validation.
- Bitcoin’s biggest security risks may increasingly sit outside its base layer.
Bitcoin did not fail when nearly $320 million in Bitcoin (BTC) was drained from the Liquid Network. Its consensus remained intact, no Bitcoin was created out of thin air, and no private keys securing the underlying blockchain were compromised.
Yet roughly 4,000 BTC still left infrastructure designed to extend Bitcoin beyond its base layer.
That distinction is what makes the Liquid exploit more consequential than another large crypto hack. The attackers found a vulnerability in Liquid’s transaction-validation software that allowed them to create unbacked L-BTC and exchange those tokens for real Bitcoin held by the Liquid Federation. Roughly 4,000 of the federation wallet’s approximately 4,200 BTC was withdrawn, representing about 95% of its reserves.
Most of the money ultimately came back. The self-described “white hats” returned 3,400 BTC after Blockstream patched affected infrastructure, while roughly 598.5 BTC, worth about $47 million at the time, remained under their control.
‼️ Self-proclaimed whitehat hackers used a vulnerability to drain Liquid Network of 4000 BTC ($320M USD) yesterday.
Just now they've returned 3400 BTC and kept 600 BTC ($47M) as a 'bug bounty'. They've also 'responsibly' disclosed the vulnerability in a PGP-encrypted message. pic.twitter.com/Ajccd7k5RR
— International Cyber Digest (@IntCyberDigest) September 7, 2026
But focusing on the recovery risks missing the larger problem. Liquid demonstrates that Bitcoin’s security does not automatically extend to every network, bridge, or financial instrument built around it.
As more BTC moves into sidechains and other layers, Bitcoin’s weakest point may increasingly sit outside Bitcoin itself.
The attack targeted the promise Behind L-BTC
Liquid, launched by Blockstream, is a federated Bitcoin sidechain designed to provide capabilities that are difficult to implement directly on Bitcoin, including faster settlement, confidential transactions, and asset issuance.
Users can lock BTC and receive L-BTC on Liquid. Under normal circumstances, every L-BTC should correspond to Bitcoin backing it.
The attackers found a way around that fundamental accounting rule.
Security company CertiK traced the vulnerability to the Elements codebase underlying Liquid and, specifically, to how the system cached verification results for cryptographic range proofs used in Confidential Transactions. These proofs allow the network to verify that hidden transaction amounts remain valid without revealing those amounts.
Because verifying them is computationally expensive, successful checks could be cached. According to CertiK, ambiguous encoding of the cache key allowed different validation inputs to produce the same cached entry.
The attackers first created transactions that caused legitimate verification results to be stored. They could then submit malicious data that pointed toward those cached results, allowing affected nodes to treat information as previously verified rather than checking it again.
3,998.5 unauthorized L-BTC was created as a result.
Chainalysis compared the basic mechanism to manipulating a bank account balance without depositing money and subsequently withdrawing the artificial balance as real cash.
The comparison captures why the exploit was so serious. The attackers did not counterfeit Bitcoin; they counterfeited a claim on Bitcoin.
SideSwap wasn’t hacked — and that’s part of the problem
Once the unbacked L-BTC existed, the attackers needed to turn it into actual BTC.
They did so through SideSwap’s legitimate peg-out service.
SideSwap said a customer sent 4,000 L-BTC to its service at 14:05 UTC. The service processed the transaction normally, burning the L-BTC with valid peg-out authorization. Twenty-three minutes later, the Liquid Federation paid 3,996 BTC to the customer’s Bitcoin address.
Critically, SideSwap’s peg-out authorization key was not compromised.
Neither were the other keys identified during the initial investigation.
That sounds reassuring until the architecture is considered more closely.
SideSwap did not need to be compromised because Liquid had already accepted the fraudulent L-BTC as valid. SideSwap said its systems had no way to distinguish the artificially created tokens from legitimately backed L-BTC.
The bridge behaved as expected based on incorrect information supplied by the layer underneath it.
That is arguably more important than a conventional key theft. Multisignature wallets and hardware security modules can protect Bitcoin against unauthorized signatures, but they cannot necessarily protect against a system incorrectly deciding that an authorized withdrawal should occur.
Bitcoin subsequently saw correctly authorized transactions and processed them.
From Bitcoin’s perspective, nothing had gone wrong.
The “white hat” label has divided security experts
The attackers’ behavior added another unusual dimension.
After withdrawing the Bitcoin, they began communicating with Blockstream through Bitcoin’s “OP_RETURN” field and identified themselves as white hats. They told developers the network remained vulnerable and said they would return most of the funds after the problem was fixed.
Blockstream eventually confirmed that affected bridge nodes had been patched. The attackers then returned 3,400 BTC in a single transaction, leaving 598.5 BTC in an address they controlled.
Not everyone in the security industry accepts the white-hat characterization.
Ledger chief technology officer Charles Guillemet questioned the description, arguing that legitimate security researchers would not normally empty almost an entire bridge before asking its developers to make contact. He compared the behavior with previous crypto exploits where attackers sought negotiations after obtaining control of funds, although he also acknowledged that conventional criminals do not typically contact victims in this manner either.
Security researcher Kirill Firsov was more explicit. He argued that responsible disclosure could have been demonstrated with a minimal proof-of-concept transaction rather than removing billions of dollars’ worth of assets, pointing to his team’s previous use of a $100 test transaction when reporting a serious vulnerability.
That criticism matters because the 600 BTC retained by the attackers has sometimes been characterized as a bounty.
There is no confirmed public agreement supporting that description. Chainalysis said that while observers speculated the remaining Bitcoin could represent a de facto reward, neither Blockstream nor the actors had publicly confirmed such an arrangement as of its analysis.
Bloomberg ETF analyst James Seyffart also highlighted the unusual economics, joking that 600 BTC was “quite the tip” while observing that victims of Bitcoin theft might willingly surrender 15% if doing so guaranteed recovery of the remaining 85%.
That may be economically rational after an exploit, but it raises an uncomfortable security question: If someone can withdraw hundreds of millions of dollars without permission and retain tens of millions after returning most of it, where does bug hunting end and extortion begin?
Liquid’s federation wasn’t enough to prevent the failure
Liquid’s design also highlights how dramatically security assumptions change once BTC moves beyond Bitcoin.
The network is governed by a federation of more than 80 organizations, while 15 rotating functionaries operate core infrastructure. Moving Bitcoin from the federation wallet requires 11 signatures.
Liquid hack explained.
Liquid has confidential transactions that hide the amounts for improved privacy. A bug in how these transactions are validated caused inflation of Liquid BTC (L-BTC) and allowed hackers to empty the entire side chain.
Liquid nodes don't see the amounts…
— calle (@callebtc) September 7, 2026
That should make straightforward private-key theft extremely difficult.
Yet the latest exploit largely bypassed that protection because the problem occurred earlier in the decision-making chain.
The federation’s signing system can answer: Has the required number of authorized parties approved this Bitcoin transaction?
It cannot necessarily answer: Was the L-BTC that caused them to approve it legitimately created?
The September exploit targeted the second question.
Former Blockstream chief strategy officer Samson Mow initially described the working theory as a node-level confidential transactions vulnerability unrelated to peg-out authorization keys or hardware security modules. After 3,400 BTC was returned, he said the network remained paused, while Blockstream and federation members implemented additional fixes, addressed a chain split, and prepared for a safe restart.
The discussion between @Blockstream and the white-hat hacker (WHH) regarding the ~4000 BTC from @Liquid_BTC is happening in public. It seems to be their preference over email. As it's hard to follow the chain of messages in OP_RETURN, here's a summary with links.
11:30 AM PDT -… https://t.co/IEXyFpBITx
— Samson Mow (@Excellion) September 7, 2026
Mow also stressed an important distinction for users: Liquid functionality could be disrupted, while normal Bitcoin transactions continued operating.
Technically, that is exactly right.
Strategically, it identifies Bitcoin’s emerging problem.
“Built on Bitcoin” doesn’t mean secured like Bitcoin
Bitcoin’s conservative architecture is frequently criticized because it cannot easily replicate everything possible on programmable blockchains.
That limitation is also part of its security model.
Adding faster settlement, confidential transfers, lending, tokenization, and complex financial markets generally requires moving functionality elsewhere. Once that happens, additional security assumptions appear.
Liquid relies on federation members and Elements software. Wrapped Bitcoin products may rely on custodians. Bridges can depend on multisignature arrangements or smart contracts. Lending protocols introduce oracle and liquidation risks. Other Bitcoin layers may introduce sequencers, validators, or alternative consensus mechanisms.
The underlying asset may still be BTC, but the route through which users gain exposure to it can contain vulnerabilities Bitcoin’s consensus cannot detect.
Liquid provides an unusually clean demonstration.
The Bitcoin network did not know whether 4,000 L-BTC had been legitimately created because validating L-BTC supply is not Bitcoin’s responsibility. Once valid Bitcoin transactions were signed by parties authorized to spend the federation’s BTC, Bitcoin enforced them exactly as designed.
Bitcoin’s security therefore protected the attacker receiving the BTC just as effectively as it had previously protected the federation holding it.
Bitcoin’s weakest link is moving up the stack
As Bitcoin’s value and hashrate increase, attacking the base blockchain directly becomes extraordinarily expensive. Infrastructure surrounding Bitcoin can offer a much softer target while still providing access to enormous pools of BTC.
Liquid had approximately 4,200 BTC in its federation wallet. A software-validation vulnerability was enough to expose 4,000 BTC without breaking Bitcoin’s cryptography or stealing the federation’s signing keys.
That asymmetry will become increasingly important if Bitcoin’s wider financial ecosystem grows.
A network can ultimately control billions of dollars in BTC while relying on a much smaller codebase, developer group, or validator set than Bitcoin itself. Attackers have little reason to challenge Bitcoin’s proof-of-work security if vulnerabilities elsewhere can provide access to the same asset.
The Liquid incident ended far better than it could have. Around 85% of the withdrawn Bitcoin was returned, developers patched the affected infrastructure, and the Bitcoin blockchain continued functioning normally.
But recovery should not become the central lesson.
LIQUID NETWORK UPDATE: OPERATIONS RESUMED, RECOVERY ONGOING
Status as of September 10, 2026, 10:00 UTCRecovery Status
The Liquid Network has entered the next phase of its controlled resumption. As a precautionary measure, block production has resumed without transactions while… https://t.co/PbyBXIzdQc
— Liquid Network 🌊 (@Liquid_BTC) September 10, 2026
The important event occurred when nearly the entire reserve successfully left the federation wallet.
Bitcoin’s next security challenge may have surprisingly little to do with breaking Bitcoin. As more financial activity moves onto sidechains, bridges and higher layers, attackers only need to compromise the systems deciding who has a legitimate claim on BTC.
Bitcoin makes actual BTC extraordinarily difficult to counterfeit.
Liquid showed that counterfeiting a sufficiently convincing claim on Bitcoin can be enough.
Unlock premium content
Create a free account to continue reading AlphaClub articles and access exclusive features.
Share


