The Ledger implant: a supply-chain attack, not a Ledger hack

For ten days the phrase “Ledger hack” has been everywhere, and it is the wrong phrase. Ledger's systems were not breached and its firmware was not broken. What happened is older and simpler: devices were physically tampered with somewhere between the factory and the buyer, and the tampering read the recovery phrase off the screen while the owner set the wallet up. This article separates what is confirmed from what is not, explains why the hardware did not save anyone, and lists what you can check right now without touching a device you no longer trust.

What is confirmed

• The timeline. Reports surfaced on 9 October 2026. On 10 and 11 October Ledger stated that one affected device contained an unauthorised hardware implant, and that every confirmed loss involved a device sold through CryptoBilis, a reseller operating in Indonesia, Malaysia and the Philippines. Ledger asked resellers in seven Asian markets to suspend sales and pull stock.
• The mechanism. According to Ledger's security team, the implant monitored the device's screen during setup, captured the recovery phrase as it was displayed, and sent it out through an embedded eSIM over LTE. It did not interact with the Secure Element or with any transaction or signing logic. It did not need to: whoever holds the recovery phrase can derive every key the device will ever generate.
• Ledger's own channels. Ledger says it has no indication that its infrastructure, its systems or its direct sales channel were compromised.
• Some funds were frozen. Tether froze about 10 million dollars in USDT linked to the thefts.

What is not confirmed

• The total. Bitquery traces about 93 million dollars from 315 wallets across six chains; other firms put it between 72 and 103 million. These are on-chain estimates of funds that left accounts protected by a Ledger device. Ledger has not published a figure and has not explicitly tied every traced loss to the implant.
• Other retailers. Mark Karpelès, formerly of Mt. Gox, says a Ledger Flex bought at Singapore's Changi Airport contained a similar module, and that reports of tampered devices go back to December 2025. A German customer reported a tampered unit and MediaMarkt suspended Ledger sales in Europe. Ledger has not linked either case to CryptoBilis, and no distributor has been shown to have taken part.
• How many devices. Nobody has said.

One detail from the investigations is worth keeping, because it says something about who gets hit: Chainalysis traced about 60 BTC, close to five million dollars, drained from three Ledger addresses belonging to the operator of a “task scam”. The attackers did not pick their victims. The implant took whatever was set up on it.

Why a hardware wallet did not help

A hardware wallet makes one promise: the private key is generated inside the device and never leaves it. The signing happens inside; the computer only ever sees a signed transaction. That promise is intact. The Secure Element in these devices did exactly what it was built to do.

The attack went around it. The recovery phrase is the one moment the root secret has to exist outside the chip, because you have to write it down. It appears on the screen, in plain words, once. An implant that watches the screen at that moment gets the same thing a thief would get by reading your backup card. From then on the hardware is irrelevant: the attacker rebuilds the wallet on their own machine and spends from there, and nothing on your device, no PIN, no passphrase entered later, no firmware check, can tell that it has happened.

This is why Ledger's advice is not “update the firmware”. It is “move everything to a fresh seed on a device you got directly”. The seed is the thing that was stolen, and a stolen seed cannot be un-stolen.

What you can check right now

If you own a Ledger bought from a third party in the last year, the honest list is short.

1. If the device is still sealed, leave it sealed. Ledger's instruction for CryptoBilis stock is not to initialise it. Do not run the setup to “see if it is fine”; the setup is the moment the attack happens.
2. Look at your addresses without touching the device. Your extended public key, the xpub, ypub or zpub that your wallet software already has, is public information by design. Paste it into a read-only tool and you see every address the wallet has used, its balance, and whether it has ever spent. ShorWatch does this in your browser against public chain data; nothing is sent to a server and the device never has to be plugged in. If the balances are where you left them, nothing has moved. If an address you never spent from shows a spend, you have your answer.
3. If anything has moved, assume the whole seed is gone. Every address under that seed is controlled by the attacker, including the ones that still show a balance. Move what is left to a new seed immediately. There is no partial compromise here.
4. If nothing has moved and the device came from a third party, move the funds anyway. A fresh seed on a device bought from the manufacturer costs a transaction fee. Keeping funds on a seed that might have been read costs the funds.
5. Buy direct, and treat a device you did not unbox yourself as a device someone else set up. The seal is not the security. The seal is a sticker.

What monitoring can and cannot do here

We sell address monitoring, so this is the place to be precise rather than flattering. An alert that fires when a watched address spends would have told the victims within minutes rather than days that something was wrong. That is useful: several people in this incident held funds across more than one seed, and minutes are what you need to move the rest. It would not have stopped the first transaction, because nothing read-only can. A tool that claims it can block a theft is describing custody, not monitoring.

For a seed theft specifically, the alert is a smoke detector, not a sprinkler. If that is the trade-off you want, it exists. If you expected more from it, you should know that before paying anyone for it.

Where this meets the quantum question

This site is about a different threat: a future quantum computer deriving a private key from a public key that is already on the chain. The Ledger implant has nothing to do with quantum computing, and we would rather say that plainly than stretch a headline.

What the two share is one discipline. In both cases the question is where is my key exposed, and when. The seed is exposed for the seconds it is on a screen; the public key is exposed from the first spend onwards, forever. The recovery from the Ledger incident, moving coins to fresh addresses on a fresh seed, happens to be the same move that protects coins at rest against the long-range quantum attack: a never-spent, hash-based address shows the world only a hash. If you are migrating anyway, it costs nothing extra to land on addresses that are protected on both counts. Our first article explains which address types those are.