In late July 2026, Coinkite—the manufacturer of Coldcard hardware wallets—confirmed that certain firmware versions generated wallet seed phrases using a weak pseudo-random number generator instead of the device’s hardware random number generator. The result was a dramatic reduction in the entropy available during private key generation. Attackers exploited this flaw to reconstruct wallet seeds outside the device and systematically drain wallets in several successive waves, with losses now exceeding 1,300 BTC—approximately USD 88–90 million—across thousands of addresses.
Coinkite released patched firmware and urged users to migrate their funds to newly generated seed phrases, expressly acknowledging that updating the firmware does not secure wallets that were originally created with the vulnerable software. Its founder publicly apologized and offered to assist victims with police reports and insurance claims. Yet none of these measures restores the stolen bitcoin. Once the funds have been transferred on-chain, they are, for all practical purposes, irrecoverable.
The Problem Is Not Merely Technical—It Is Structural
For years, the industry has debated how to generate wallet seeds securely. Far less attention has been given to what happens when that process fails—not because of user negligence, but because of a manufacturer’s defect—and the result is an objectively identifiable theft, with a known technical cause and clearly identifiable victims.
Today, the architecture of cryptocurrency itself provides only one answer: nothing.
The immutability of blockchain and the absence of intermediaries—core principles that give Bitcoin its strength—become, in circumstances like these, the very obstacle that prevents any effective remedy, even where there is broad technical consensus regarding the source of the vulnerability and the manufacturer itself has acknowledged the defect.
BACS’ Proposal: A Legal Oracle as a Layer of Remedy, Not of Control
At BACS, we have consistently argued that the answer cannot simply be better auditing of key-generation code—although that remains essential. Recent history has shown that vulnerabilities will continue to occur, as illustrated by Milk Sad in 2023, Ill Bloom in early 2026, and now the Coldcard incident.
What is also needed is a second layer: a legally governed recovery mechanism embedded within the protocol or token standard, capable of being activated only following a verifiable arbitral award or court decision.
In practice, this means implementing a legal oracle: a component capable of certifying on-chain that a recognised arbitral authority—such as BACS—has determined that specific funds originate from a technically proven theft, and authorising, within a strictly defined time window and subject to full transparency guarantees, a narrowly tailored recovery action, such as freezing or redirecting those assets.
This is not about introducing a backdoor or granting discretionary control over users’ funds.
It is about creating an exceptional, transparent and auditable mechanism that operates exclusively through a formal legal procedure offering the same safeguards as any arbitral process: adversarial proceedings, expert technical evidence, reasoned decisions and due process. It would apply only in exceptional cases—such as this one—where the unlawful origin of a transaction can be established objectively.
Why Coldcard Is the Perfect Case Study
The Coldcard incident illustrates precisely the type of situation this mechanism is intended to address.
There is a manufacturing defect acknowledged by the manufacturer itself, a theft pattern independently documented by security researchers, and thousands of victims with virtually no realistic avenue to recover their assets other than years of cross-border litigation against a company whose civil liability—even in light of the available evidence—is already becoming the subject of legal debate.
Had such a mechanism existed, the central legal question would not have been whether Coinkite could ultimately be held civilly liable. Instead, the focus would have been whether the theft satisfied the objective criteria required to trigger the recovery mechanism.
That is a far simpler and faster question to answer—and one that does not require years of international litigation before victims can obtain relief.
A Call to Action
BACS invites hardware wallet manufacturers, protocol developers and token issuers to evaluate the incorporation of legal-oracle-based recovery mechanisms into future blockchain architectures.
We are available to work with the industry in developing the necessary arbitral framework—including activation criteria, evidentiary standards, procedural safeguards, transparency requirements and time limits—to make this type of solution both legally robust and technically viable, without compromising the permissionless nature of public blockchain networks.
