LIVE MARKET DATA MON 10 AUG 2026 UTC [ VIEW ALL COINS ]
// Bitcoin

Coldcard Entropy Flaw Linked to 594 BTC Swept From ~500 Single-Sig Wallets

A firmware bug cut Coldcard seed entropy from 128 to 72 bits; on-chain reports tie it to a 594 BTC sweep across single-signature wallets on July 30-31.

James Corrigan · ·3 min read
Coldcard Entropy Flaw Linked to 594 BTC Swept From ~500 Single-Sig Wallets

Roughly 594 BTC moved out of approximately 500 single-signature wallets on July 30 and 31, 2026, in what validated incident notes tie to a seed-generation flaw in Coldcard firmware. The bug reportedly cut entropy from the intended 128 bits down to 72 bits by substituting a predictable software routine for the device’s hardware random number generator.

Which devices and firmware are exposed

The affected range spans Coldcard Mk3 firmware 4.0.1 through 5.0.3, Mk4 and Mk5 units running firmware prior to 5.6.0, and Q devices before version 1.5.0Q. Fixed builds — 5.6.0 for Mk4/Mk5 and 1.5.0Q for Q — are already referenced in the incident notes, narrowing the exposure window to devices that have not been updated.

Not every seed generated on those firmware versions is treated as compromised. Under the validated notes, seeds strengthened with a BIP-39 passphrase, or generated using at least 50 dice rolls rather than relying solely on the device’s onboard entropy, fall outside the at-risk category.

Why 72-bit entropy breaks the security model

Bitcoin’s seed-phrase model relies on a search space too large to brute-force at 128 bits of entropy. Dropping to 72 bits shrinks that space dramatically, turning what should be a computationally impossible guess into a narrower, potentially tractable target for an attacker with sufficient resources.

That distinguishes this incident from a routine firmware bug. A display glitch or a signing error creates transaction-level risk that a careful user can catch. A weak entropy source undermines the wallet at the point of creation — meaning a user who has never leaked a key, clicked a phishing link, or exposed a private key could still be exposed if their seed was generated under the affected code.

Single-signature exposure versus multisig resilience

The reported sweep concentrated on single-signature wallets, where one seed controls all funds and there is no secondary approval layer to slow an attacker who can derive or guess that seed. Multisignature setups don’t eliminate every wallet-level failure mode, but they reduce the blast radius of a single compromised seed by requiring additional keys to authorize a transaction.

The event reinforces why serious custody stacks increasingly combine multisig, BIP-39 passphrases, dice-derived entropy, geographically separated backups and hardware from multiple vendors — not because every holder needs institutional-grade setups, but because Bitcoin has no chargeback mechanism once coins move on-chain.

What traders and holders should check

For holders running Coldcard devices, the operative question isn’t simply which model they own, but which firmware version generated their seed and whether that seed relied on device-only entropy or was reinforced with a passphrase or dice rolls. Updating to firmware 5.6.0 or 1.5.0Q closes the vulnerability going forward, but funds already secured under weak-entropy seeds may warrant migration to freshly generated wallets regardless of the update.

With roughly 594 BTC already reported swept, the incident is a reminder that hardware wallets shift custody risk rather than eliminate it — verification of firmware provenance and entropy sourcing remains the user’s responsibility, not a solved problem handled entirely by the device.

Sources

More Bitcoin