0x0bbafd6d0bba…0bbafd6a
Coinkite Says Firmware Flaw Affected 0.17% of Coldcard Users
Coinkite CEO Rodolfo Novak says a firmware flaw affected about 0.17% of Coldcard users, while 1,830 BTC was stolen from roughly 9,200 addresses in the cited incidents.

Outputs
Coinkite CEO Rodolfo Novak said a firmware flaw affected about 0.17% of Coldcard users.
About 1,830 BTC, worth roughly $116 million at the time, was stolen from around 9,200 addresses.
Among 256 Coldcard victims who sought help, 245 had 2,585 addresses tied to 895 BTC.
Coinkite has not disclosed a user count or sales figures and is preparing a technical review.
The first reported Coldcard theft incident was July 30; no year was specified.
By Nathan Brooks
About 1,830 BTC, valued at roughly $116 million at the time, was stolen from approximately 9,200 addresses in the reported incidents. Coinkite CEO Rodolfo Novak said a firmware flaw affected “about 0.17%” of Coldcard users.
Novak did not give a user count, and Coinkite has not confirmed how many people were affected. The company also has not disclosed sales figures. At 0.17%, the reported share is equivalent to roughly one affected user in every 588 Coldcard users.
The ratio cannot be converted into an absolute victim total from the information released. The missing denominator is consequential because it prevents an independent check of the estimate against the installed base.
The theft totals and the user estimate also use different units. Roughly 9,200 addresses are not equivalent to 9,200 Coldcard devices or users, and the available information provides no one-to-one mapping.
Coinkite's disclosure does not identify affected hardware revisions, firmware versions, exploitation mechanics or mitigation steps. As a result, the 0.17% figure cannot by itself establish whether a particular device belongs to the affected population.
Among 256 Coldcard victims who sought help, 245 had 2,585 addresses tied to 895 BTC. That subset represents about 49% of the 1,830-BTC total and about 28% of the 9,200 addresses.
The comparison offers a scale check, not a complete victim ledger, and the disclosure gives no breakdown for the other 11 people in the 256-victim group. It also shows why address counts should not be substituted for user counts: the sample links more than 2,500 addresses to fewer than 256 cases.
The first reported Coldcard theft incident was July 30. The material does not specify a year, and the date does not establish when the firmware flaw was introduced or first exploited.
Novak previously described losses linked to the flaw, but Coinkite is still preparing a technical review of the incident. Until that review is complete, the 0.17% estimate and the theft totals should be treated as provisional management and incident figures rather than a completed forensic attribution.
What does this change for Coldcard users?
For users, the immediate change is evidentiary rather than a new operating instruction. The available disclosure gives no affected model list, firmware version, remediation step or review deadline.
The 0.17% estimate alone does not establish whether a particular unit is implicated, while the disclosed figures do not let owners in the assistance group reconcile the company-wide percentage with 895 BTC tied to 2,585 addresses. The scale remains material regardless: assets worth about $116 million at the time of the theft were involved, even if the reported affected-user share was comparatively small.
For Coldcard owners and prospective buyers, the incident demonstrates why a headline percentage does not replace product-specific disclosure about affected versions, exposure and corrective action. Without those details, buyers cannot use the percentage to compare device risk or decide whether a unit requires remediation.
Coinkite has not announced a deadline for the technical review. If the review ties the defect to a defined set of devices and versions, it will provide a concrete basis for remediation and exposure analysis.
Until then, the affected-user total remains unconfirmed and the reported $116 million valuation remains broader than the victim-assistance subset. The next firm checkpoint is publication of the review, although no date has been set.
via tokenpost.com (Source)
More from Nathan Brooks
Show full bio
Market editor covering money, funding and market shifts around crypto and web3 industry at Mempool Brief.
1,512 articles