BIP-110 Activation Frozen After Coldcard Exploit: Is the Soft Fork Dead?

Bitcoin developers have frozen activation of BIP-110, a proposed soft fork to restrict transaction data, after a critical vulnerability in Coldcard hardware wallets exposed user seed phrases to theft. The pause underscores how security crises in one part of the ecosystem can derail consensus-layer upgrades and exposes deep rifts over Bitcoin’s blockspace policy among miners, operators, and institutions.

• BIP-110 activation delayed indefinitely by developers citing Coldcard exploit that compromised seed phrase randomness in wallets since March 2021
• Mining support collapsed to 2.63% of blocks in current voting period, far below 55% threshold needed and only one-twentieth of required support
• Soft fork design requires node operators to enforce rules voluntarily; freeze signals developer uncertainty about whether community will switch back from BIP-110 nodes

• 2.63% Mining pool support for BIP-110 in current period versus 55% threshold needed
• 72 bits Randomness used in Coldcard seed generation instead of promised 128 bits
• December 1, 2025 Date mining voting on BIP-110 began before activation was frozen

Bitcoin developers have paused the activation of BIP-110, a proposed soft fork designed to constrain the amount of arbitrary data stored in Bitcoin transactions, after a serious vulnerability surfaced in Coldcard hardware wallets exposed users to seed phrase theft.

Udi Wertheimer, a lead voice in the BIP-110 effort, announced the freeze on social media, citing “the coldcard incident” and asking node operators running BIP-110 software to revert to standard Bitcoin nodes without specifying a new activation timeline.

The move reflects a critical juncture for Bitcoin’s technical governance: a single security failure in one wallet manufacturer has halted a rule change that had already struggled to attract miner support, raising questions about whether consensus-layer upgrades can survive external shocks when community backing remains fragile.

Coldcard’s Seed Generation Flaw Exposed 72-Bit Entropy Instead of 128-Bit Standard

On July 30, 2025, Coinkite disclosed that its COLDCARD hardware wallets had generated seed phrases, the master cryptographic keys controlling user funds, using far less randomness than advertised. The wallets created seeds with approximately 72 bits of entropy instead of the standard 128 bits, a shortfall that makes seeds vastly easier to brute-force.

Firmware released since March 2021 was affected, meaning a significant portion of COLDCARD’s installed base faced exposure, though the company has not quantified how many wallets or how much value was compromised.

Updating a device to patched firmware does not restore a seed phrase that was already generated incorrectly, users must move their funds to new, properly seeded wallets or risk permanent loss.

Coinkite has instructed all COLDCARD owners using affected firmware to migrate their holdings immediately. Evidence already emerged of thieves exploiting the flaw to drain wallets tied to weak seeds. The incident arrived at precisely the moment when Bitcoin’s technical community was voting on BIP-110, a rule change already contentious and fragile.

For institutional investors managing custody or mining infrastructure, the timing underscores a broader risk: hardware wallet vulnerabilities can propagate far faster than consensus-layer fixes, and market pressure to address security crises can override technical merit in determining which upgrades proceed.

Mining Support Never Reached 20% of Threshold Before Developers Froze Activation

BIP-110 has required 55% of blocks mined during a two-week voting window to signal support, equivalent to 1,109 blocks, since mining began on December 1, 2025. As of the freeze announcement, the proposal had secured only 30 blocks flagging approval, roughly 2.63% of the 1,068 blocks mined in the current voting period.

This marked BIP-110’s strongest showing to date; every prior two-week window since December had fallen below 1.3%, leaving the soft fork more than twenty times short of the threshold needed to proceed.

With approximately 948 blocks remaining before the end of the current voting window, mathematical passage was already impossible, even unanimous support from the remaining blocks would yield only about 48% backing.

Michael Saylor, the prominent Bitcoin holder and MicroStrategy chair, has suggested that nearly all BIP-110 support originates from a single mining pool, raising questions about whether the votes reflect genuine consensus or concentrated operator preference.

Adam Back, chief executive of Blockstream, has publicly flagged the 55% threshold itself as dangerously low for a soft fork, citing chain-split risk if nodes enforce incompatible rule versions.

The voting mathematics had collapsed days before developers announced the freeze, meaning the pause was not a surprise tactical response but a formal acknowledgment of consensus failure already evident on-chain.

Soft Fork Enforcement Depends on Node Operators Switching Back, Not Miners Alone

BIP-110 functions as a soft fork, meaning it tightens Bitcoin’s rule set rather than expanding it. Unlike a hard fork, which requires universal adoption to avoid chain splits, a soft fork theoretically allows non-upgraded nodes to coexist with upgraded ones, but only if upgraded nodes enforce incompatible rules separately.

A second enforcement phase was scheduled to begin at block 961,632, approximately six days after the freeze announcement, when BIP-110 nodes would begin rejecting any block that did not signal support. At that point, nodes running BIP-110 code would fork away from the rest of the network unless the broader Bitcoin ecosystem upgraded or the proposal died.

Wertheimer’s plea for BIP-110 node operators to revert to standard software highlighted the critical vulnerability in soft fork activation: no single entity controls Bitcoin’s rules, and no mechanism exists to force nodes to downgrade once they have upgraded. The freeze is fundamentally a request, not a command.

Whether operators comply will determine whether BIP-110 truly collapses or lingers as a fragmented minority fork. For institutions running nodes, the pause creates operational ambiguity, maintaining BIP-110 software while the community fractures carries chain-split risk, but reverting signals investment in a failed initiative.

Institutional investors and mining operators now face a window of vulnerability as nodes decide whether to downgrade. If significant hash power or node infrastructure ignores the freeze and enforces BIP-110 rules at block 961,632, Bitcoin could temporarily split into two versions.

Even if that split resolves quickly in favor of non-BIP-110 nodes, the event would damage confidence in Bitcoin’s upgrade process and likely set back future soft forks for months.

The Blockspace Debate Remains Unresolved as BIP-110 Backers Regroup

BIP-110 was designed to address what supporters view as “blockspace spam”, transactions carrying data unrelated to payments that clutter the Bitcoin ledger and increase fees. The soft fork would impose a one-year limit on how much arbitrary data could be embedded in each transaction.

Critics, including privacy advocates and decentralized finance developers, argue that Bitcoin should not police what users store on-chain and that data-bearing transactions have legitimate uses beyond spam.

The fundamental disagreement over Bitcoin’s purpose, settlement layer versus open data platform, predates BIP-110 and will outlast its freeze. Proponents believe constrained blockspace justifies the restriction; opponents see it as mission creep.

The poor miner voting results suggest neither camp achieved consensus, but the Coldcard incident gave developers an exit ramp to postpone the fight without resolving it. Wertheimer stated the delay reflects timing, not doubt in BIP-110’s merit, but his statement came without a new activation date, leaving the proposal in indefinite limbo.

For institutional participants, the unanswered question is whether the freeze represents a tactical pause during a security crisis or the effective end of BIP-110 as a viable upgrade path.

The next critical moment will arrive when developers either announce a new activation timeline or formally withdraw BIP-110. Any announcement will signal how seriously Bitcoin’s technical leadership takes miner opposition and whether external security events can halt consensus-layer upgrades. Institutional node operators should monitor whether significant hash power dow