A fight between Ledger’s CEO and the former boss of Mt. Gox has turned a hardware-wallet security incident into a wider argument about trust.

In an October 17 exchange on X, Mark Karpelès challenged how prominently Ledger had warned customers about spying implants hidden inside genuine devices. Pascal Gauthier defended the company’s work, then brought up his critic’s history.

“Coming from Mt Gox CEO… Come on man. You are no white knight. Get real.”

— Pascal Gauthier, CEO, Ledger, as reported by crypto.news

That was Gauthier’s response, as reported by crypto.news. The remark made the dispute personal. The underlying question is more consequential: how should a wallet maker explain a threat that its authenticity check cannot fully detect?

The device can be genuine and still be compromised. In its October 15 response, Ledger said criminals acquired legitimate devices, added hidden components capable of capturing recovery phrases during setup, and returned the altered hardware to distribution through resellers in Southeast Asia.

Ledger described an attack on devices after they left its control. It said its chip, operating system and servers had not been compromised.

“This was an attack on the distribution channel after the devices left our control.”

— The Ledger Team, October 15 company statement

The distinction matters because an authentic security chip does not establish that every other component inside a device is untouched.

Ledger’s own buying guide explains the limitation. Its Genuine Check verifies the identity of the Secure Element—the protected chip at the heart of the device. If that original chip remains intact, unauthorized additions can escape the check.

For buyers, a successful authentication result therefore answers a narrower question than “has anyone physically modified this wallet?”

Karpelès also corrected his original criticism. He acknowledged that Ledger had published an Academy article addressing implant risks, retracting his assertion that the company had said nothing about them. His remaining argument was that the warning deserved greater prominence.

The guide lists March 6 as its publication date and September 18 as its latest update. Its present contents cannot, by themselves, establish when every photograph or warning first appeared. Nor does knowledge of this general attack method establish prior knowledge of the specific compromised devices.

Reports have put suspected losses above $90 million, but Ledger has not confirmed that as a final verified total. Karpelès’s suggestion that clearer warnings could have prevented those losses remains a claim, not an established finding.

Ledger is changing how retailers handle its products. The company said it asked authorized resellers and distributors to suspend sales in Southeast Asia and Hong Kong. It also reinforced its prohibition on reselling returned devices and introduced a requirement to keep stock off shelves and displays, handing it over at purchase.

Those measures address a physical part of digital ownership: what happens between a product leaving its maker and reaching its user.

For Blockster readers, that is the wider significance. As AI agents seek access to financial accounts, understanding exactly what a security control verifies becomes increasingly important. Device authenticity, permission to transact and protection of recovery credentials are separate responsibilities.

Self-custody gives users direct control, but its promise depends on understandable safeguards. A customer should be able to tell what a successful check proves, what remains outside its scope and where to find current incident guidance.

Ledger’s documentation also warns against entering a recovery phrase into a website, phone or computer application. That remains a separate phishing risk; avoiding it does not resolve the physical implant problem.

The industry’s challenge is to make those boundaries clear before customers have to learn them through a public argument.

Reporting by Eric Spivak

LedgerSecuritySelf-CustodyWallets