BIP-110 mines two blocks, then stalls: a Bitcoin soft fork the network is refusing to lift

BIP-110, the user-activated soft fork proposal that has lingered on Bitcoin's activation calendar for months, produced two blocks on a breakaway chain early on 9 August 2026 before going quiet, according to CoinDesk's coverage of the activation window. The chain inherited bitcoin's mainnet difficulty with only a sliver of the network's hashpower, leaving blocks hours apart while the two networks continued to accept the same transactions.

What looked, on paper, like an attempted governance test of bitcoin's decentralised change process is resolving, at least for now, as a non-event: two proof-of-work blocks of contested validity, broadcast into a vacuum, with miner support measured in low single-digit percentages and the spot price unmoved on the news. The episode crystallises a question the protocol community has been chewing on since BIP-110 entered mandatory signalling: in a system where nodes, not miners, can dictate activation, what does "consensus" actually mean when almost nobody is mining the new rules?

Two blocks, one chain, almost no hashrate

The breakaway chain produced its first block on 9 August 2026 at roughly 05:08 UTC, according to CoinDesk's running tally of BIP-110's activation attempt. A second block followed at a multi-hour interval, in line with the difficulty the fork inherited from bitcoin mainnet. With under 3% of miner signalling recorded ahead of the activation milestone on 8 August, blocks on the minority chain took far longer to find than on the chain the rest of the network continued to extend.

Cointelegraph's coverage on 8 August 2026 framed the moment as a test of whether enforcing nodes alone can sustain a soft-fork change against limited miner participation. The data point the early hours of 9 August supplied is provisional but pointed: two blocks mined does not a chain make. A network that takes hours to confirm a block is, functionally, a network that is not confirming transactions in any normal sense, and one in which miners are rationally withholding work until they can tell which side the rest of the economy will transact on.

The price reflected none of this. Investing.com's coverage on 8 August recorded a push toward $65,000 even as the fork window opened.

Why BIP-110 refuses to die

The puzzle of BIP-110 has never been its technical content alone. CoinDesk's explainer on 6 August made the structural point plainly: the proposal has attracted only a sliver of miner support, yet its user-activated design means it continues marching toward its activation date regardless of where the hashpower sits. In a protocol where node-runners can encode rules into their consensus logic and miners can simply choose not to extend blocks under those rules, "activation" and "adoption" are two different things, and the gap between them is the entire political fault line.

Developer warnings during the window sharpened the operational stakes. CoinDesk reported on 8 August that holders who sell coins produced on a BIP-110 minority chain could find their transactions replayed on bitcoin itself, putting their mainnet bitcoin at risk in the process; doing nothing, until chains can be cleanly separated, was described as the safer move by the developers quoted. The replay risk is the kind of detail that turns an abstract governance debate into a concrete wallet-level decision for anyone holding bitcoin through the window.

Self-custody, hardware wallets, and the other story underneath

While the protocol world watched BIP-110's two-block stutter, a parallel infrastructure shock landed in the hardware-wallet world. Moneyweb's coverage on 9 August described a software vulnerability that allowed hackers to drain bitcoin from affected hardware wallets, reigniting the perennial self-custody debate inside a community whose foundational pitch is "not your keys, not your coins."

Moneyweb's framing captures the contradiction sharply: the tools marketed as the safest way to hold bitcoin, free of exchange counterparty risk, depend on software stacks with their own attack surface. The vulnerability, as Moneyweb reports it, exploited precisely the layer at which users are told their assets are most secure. Read against BIP-110's governance stress, the two stories rhyme. A network that asks its users to make their own security decisions, and a community that asks its users to make their own consensus decisions, both require a baseline of trust in code that has, this week, looked shakier than usual.

Monexus analysis: the governance test that never arrived

Monexus analysis: the interesting question was never whether BIP-110 would activate, but what its activation would measure if it did. The answer, after two blocks and a near-empty mining pipeline, is that "node-activated" carries less weight than its proponents hoped when the rest of the system declines to follow. Two blocks broadcast into an empty room tell the rest of the economy that the economic majority, miners included, is unmoved. That is a soft signal, not a hard one; nothing on-chain has been irreversibly split, and the chain's operators retain the option to wind BIP-110 back down. But the signal is there for anyone reading it.

The forward read, based on the available evidence: a successful user-activated soft fork requires either a much larger node majority willing to enforce rules that miners ignore for an extended window, or an economic actor with the capital to mine alone until the rest of the market catches up. Neither condition appears to be met in the available reporting. This publication's expectation is that the next 48 to 72 hours will resolve the question by attrition: if BIP-110's chain continues to produce blocks at multi-hour intervals and exchanges decline to credit fork-coin balances, the episode will drift into the same archive as previous near-misses.

A separate, second-order observation sits beneath the fork story. CoinDesk reported on 6 August that a volunteer group running AI models against bitcoin codebases has flagged 85 critical bugs in an "extremely bad" situation, averaging roughly one critical bug per hour per person at about $10,000 a day in compute. The hardware-wallet vulnerability reported by Moneyweb on 9 August looks, against that backdrop, less like an isolated incident and more like one entry in a category. A protocol whose security story depends on code review by an ever-smaller volunteer cohort, and on user vigilance at the wallet layer, is exposed in ways that fork politics do not address.

Desk note: Monexus framed BIP-110 as a governance stress test, not as a market-moving event; the wire coverage on 8 and 9 August leaned into the activation milestone and the price reaction, while our read puts the weight on the gap between node signalling and miner behaviour and on the adjacent self-custody story.

Wire provenance

This editorial synthesis draws on the following public wire/social posts:

• https://www.coindesk.com/tech/2026/08/09/controversial-bitcoin-fork-bip-110-mines-two-blocks-then-stops
• https://cointelegraph.com/news/bitcoin-bip-110-mandatory-signaling
• https://www.coindesk.com/tech/2026/08/07/frame-bitcoin-s-bip-110-enters-mandatory-signaling-with-less-than-3-miner-support
• https://www.coindesk.com/tech/2026/08/08/bitcoin-holders-risk-losing-real-btc-if-they-sell-coins-from-bip-110-fork-says-developer
• https://www.coindesk.com/tech/2026/08/06/why-bitcoin-s-bip-110-refuses-to-die-despite-near-zero-miner-support
• https://www.moneyweb.co.za/moneyweb-crypto/bitcoin/hack-of-supposedly-safe-bitcoin-tool-tries-faith-of-the-devoted/
• https://www.coindesk.com/tech/2026/08/06/bitcoin-developers-flag-85-critical-bugs-in-an-extremely-bad-situation
• https://www.investing.com/news/cryptocurrency-news/bitcoin-rises-toward-65000-as-fresh-security-risks-hit-infrastructure-4847628