Ledger confirmed that at least one device connected to reported wallet drains contained an unauthorized hardware implant. The confirmed cases were linked to devices sold through Southeast Asian reseller CryptoBilis, but the company has not publicly established exactly how the implant worked or verified a total loss figure.
2
4
7
What Ledger confirmed
Reports of drained wallets emerged on October 9, 2026. Ledger confirmed finding an implant in an affected user’s device the following day. In its October 11 update, the company said all confirmed cases in the reported incident involved devices sold through CryptoBilis.
2
4
7
Ledger described the number of affected devices as “limited,” without giving a verified count. It also said its own systems and direct sales channels were not compromised. Those statements narrow the investigation’s reported scope; they do not establish that every suspected theft came from an implanted device.
2
5
6
Loss estimates are not a confirmed Ledger total
Independent estimates varied as the investigation developed: early reporting put suspected losses above $86 million, while later estimates exceeded $93 million. One estimate put the losses at about $93.4 million; another reported $92.9 million across 311 wallets. Ledger has not validated these totals.
7
11
13
These estimates concern traced wallets and funds, not a confirmed count of tampered devices. The number of wallets in an on-chain analysis should not be treated as the number of affected hardware wallets.
13
14
What is known about the implant
Mark Karpelès reportedly found an extra circuit board in a Ledger device obtained in Malaysia. Accounts of his discovery describe a microcontroller, LTE hardware and an e-SIM. The proposed mechanism is that the board could read recovery words displayed during setup and transmit them over a cellular connection.
18
25
That remains a reported technical explanation, not a method Ledger has officially confirmed. Ledger confirmed an unauthorized implant was present in one examined device, but the available reporting does not establish precisely how it operated or prove that the implant caused every reported loss.
2
9
Reported precautions for CryptoBilis buyers
Ledger asked CryptoBilis to pause sales and shipments while the investigation continued; reporting also said the reseller halted its hardware-wallet sales.
4
7
For devices bought through CryptoBilis in the preceding 90 days, reported guidance was:
- Not yet initialized: Do not set up the device while the investigation is ongoing.
17
20
- Already initialized: Consider moving assets to a new signing device set up with a newly generated recovery phrase. Restoring the same potentially exposed phrase on replacement hardware would not address exposure of that phrase.
17
26
The 90-day window is a precautionary cutoff, not evidence that devices bought earlier are safe. Reports that many traced wallets first received funds in June or later do not, by themselves, establish when the devices were bought or whether they were tampered with.
What remains unresolved
The public evidence cited here does not establish how many physical devices were affected, whether every suspected loss involved an implant, or the implant’s confirmed operating method. The loss figures remain independent estimates rather than a final total verified by Ledger.
2
7
13