what Bitcoin’s failed anti-spam fork reveals about governance in 2026

The minority chain mined two blocks in eight hours, then froze. With 99.85% of hashpower on the original chain, the Reduced Data Temporary Softfork is the most decisive governance rejection since SegWit2x. Its backers are now talking about replacing the miners entirely.

Bitcoin’s latest governance crisis lasted about eight hours. On the morning of August 8, 2026, nodes running the BIP-110 soft fork rejected a block at height 961,632 that lacked the required signaling flag and began building an alternate chain. That chain produced exactly two blocks. Then it stopped.

The main Bitcoin network kept moving at its usual ten minute pace, indifferent. By the time most American traders checked their phones on Saturday morning, the split was already over in every practical sense. The BIP-110 branch had inherited the full network difficulty with almost none of the hashrate, meaning its next difficulty adjustment was estimated at roughly 350 days away. Two blocks in eight hours, then silence.

What makes BIP-110 worth examining is not its failure. Plenty of proposals fail. It is what the failure reveals about how Bitcoin governance actually works in 2026, and what happens when a faction decides the governance system itself is the problem.

What BIP-110 proposed

The Reduced Data Temporary Softfork was designed to restrict arbitrary non-financial data embedded in Bitcoin transactions. Seven rules would have capped most new outputs at 34 bytes, OP_RETURN at 83 bytes, and data pushes at 256 bytes. The restrictions were temporary, expiring after 52,416 blocks, roughly one year.

The target was explicit: Ordinals inscriptions, BRC-20 tokens, Runes, and large Taproot data payloads. Supporters argued these uses constitute spam that bloats the blockchain, drives up transaction fees for ordinary users, and burdens node operators with storage costs that serve no monetary purpose. Luke Dashjr, one of the most senior Bitcoin Core developers, backed the proposal publicly and has argued for years that non-financial data on Bitcoin is an abuse of the protocol.

The activation threshold was set at 55% of blocks in a difficulty period, or 1,109 out of 2,016. This was a deliberate departure from Bitcoin’s traditional 95% threshold, an acknowledgment by the proposal’s authors that overwhelming consensus was unlikely.

Why the miners said no

When the mandatory signaling window opened on August 7 at block 961,632, support stood at 2.53%. That is 51 blocks out of 2,016. Not 51%, but 51 blocks.

The gap between 2.53% and 55% is not a negotiation failure. It is a statement. Mining pools representing the vast majority of Bitcoin’s hashrate looked at BIP-110 and concluded that restricting fee paying transactions was against their economic interest and their understanding of how Bitcoin should work.

This is not surprising when examined through the lens of miner economics. Ordinals and Runes transactions pay fees. During peak inscription periods, those fees have pushed average transaction costs above $20, generating significant additional revenue for miners. BIP-110 proposed to eliminate that revenue for a year. No major mining pool publicly supported the proposal in the weeks leading up to activation.

The economic argument, however, only explains part of the rejection. The philosophical argument was equally decisive.

Saylor’s 110 reasons and the neutrality principle

Michael Saylor, the executive chairman of Strategy and the most visible institutional Bitcoin advocate, published a 110 point essay titled “110 Reasons BIP 110 Is a Bad Idea.” The core argument compressed into a single sentence: Bitcoin’s value comes from its neutrality, and the moment you start deciding which types of valid, fee paying transactions are acceptable, you open a door that cannot be closed.

“BIP 110 turns a spam dispute into a consensus change,” Saylor wrote, calling the precedent “extremely dangerous.” His reasoning was that if the network can reject inscriptions today, it can reject other transaction types tomorrow. The category of “non-financial data” is not self-defining. A multisignature transaction embedding metadata, a timestamp proof, or a sidechain anchor could all be classified as non-financial by future proposals using the same logic.

Adam Back, Blockstream’s co-founder and a figure whose hashcash proof of work system directly influenced Bitcoin’s design, joined Saylor in opposition. Back dismissed censorship claims from BIP-110 supporters and warned that enforcing disputed rules without broad support was the greater threat to Bitcoin’s integrity.

The weight of these voices matters. Saylor’s company holds more than 500,000 bitcoin. Back invented a core component of Bitcoin’s consensus mechanism. When both conclude that a proposal threatens Bitcoin’s foundational properties, the signal to miners and node operators is unambiguous.

The replay attack problem nobody solved

BIP-110 shipped without replay protection. This is a technical detail with serious practical consequences.

When a blockchain splits, any transaction valid on one chain may also be valid on the other. Without replay protection, a user spending bitcoin on the main chain could have that same transaction replayed on the BIP-110 chain, or vice versa. The result is an unintended transfer on the chain the user did not mean to interact with.

Bitcoin Cash, the most significant historical Bitcoin fork, included replay protection from the start. BIP-110 did not. Its designers argued that as a soft fork rather than a hard fork, replay protection was unnecessary because the rules were a subset of existing consensus. In practice, the chain split created exactly the conditions where replay attacks become possible.

Start9, the Bitcoin node manufacturer, published guidance urging users to take no action, avoid splitting coins, and wait for the minority chain to die. That advice proved correct within hours. But the absence of replay protection in a proposal that its authors knew might trigger a split reflects a governance failure within the BIP-110 project itself.

The blocksize war echo

Bitcoin has been here before. The blocksize war of 2015 to 2017 consumed years of development energy, produced Bitcoin Cash as a permanent fork, and nearly resulted in the SegWit2x hard fork that would have doubled the block size limit. That fork was called off at the last minute when it became clear that economic nodes, the exchanges, wallets, and businesses that users actually rely on, would not follow.

The parallel to BIP-110 is instructive but not exact. The blocksize war was a dispute over scaling philosophy. BIP-110 is a dispute over what belongs on the blockchain at all. Both involve a minority attempting to force a consensus change over the objection of the majority. Both failed.

But the blocksize war took two years. BIP-110 took eight hours. The speed of the resolution says something about Bitcoin’s governance maturation. In 2017, the question of whether miners or node operators controlled consensus was genuinely unresolved. In 2026, it is settled. Hashpower alone cannot force a rule change. Economic nodes must agree. And when 97.47% of miners and virtually all major exchanges, wallets, and infrastructure providers reject a proposal, the outcome is not in doubt.

SegWit itself was activated through BIP 148, a user activated soft fork that bypassed resistant miners. The lesson the community drew was that users running nodes, not miners, are the final arbiters of consensus. BIP-110’s failure is the same lesson applied in reverse: a small group of nodes and miners cannot impose rules the rest of the network rejects.

The proof of work pivot

The most remarkable development came after the fork failed. Roughnecks, a mining operation supporting BIP-110, announced it would resume mining the stalled chain and continue until a “sensible proof of work change” could replace the mining pools that refused to follow.

This is an extraordinary statement. Changing the proof of work algorithm means abandoning SHA-256, the hashing function that every existing Bitcoin ASIC is designed to compute. A new algorithm would make the entire existing mining infrastructure worthless for that chain. The minority branch would need to attract miners running different hardware, build a new difficulty adjustment schedule, and convince exchanges and wallets to list what would be, in every meaningful sense, a new cryptocurrency.

The precedent is Bitcoin Gold, which switched from SHA-256 to Equihash in 2017 to allow GPU mining. Bitcoin Gold trades at roughly $1 today. The market’s judgment on proof of work changes that split from the main chain has been consistent and severe.

Whether the BIP-110 backers follow through on the proof of work switch is an open question. But the fact that they are discussing it reveals something important: the dispute over inscriptions has not been resolved by the fork’s failure. The faction that believes non-financial data should be excluded from Bitcoin still exists. They have simply concluded that Bitcoin’s existing governance system will not give them the outcome they want.

What this means for Bitcoin’s future upgrades

BIP-110’s failure has implications beyond the inscription debate. Bitcoin has several pending proposals that would require soft forks, including OP_CTV for covenants and various vault designs. Each of these would need to clear the same governance gauntlet that BIP-110 could not.

Saylor went further than opposing BIP-110 specifically. He argued that Bitcoin’s consensus rules should be treated as a constitution, and that changes should be “rare” and never serve any faction’s convenience. This framing, if it becomes the dominant view, makes any soft fork significantly harder to activate. The threshold is no longer technical consensus but something closer to constitutional amendment, requiring not just majority support but near unanimity.

The practical effect is that Bitcoin’s protocol is becoming more difficult to change by design and by culture. Whether this is a strength or a vulnerability depends on one’s view of what Bitcoin needs. If the current protocol is sufficient for its role as a monetary network, ossification is a feature. If Bitcoin needs new capabilities to remain competitive with newer blockchains, ossification is a risk.

BIP-110 did not answer this question. But it demonstrated that in 2026, the bar for changing Bitcoin’s rules is higher than it has ever been.

What would change this analysis

If the BIP-110 minority chain successfully implements a proof of work switch and attracts meaningful hashrate, the fork becomes a permanent split rather than a failed experiment. Watch for exchange listings. If any major exchange lists the BIP-110 chain as a tradeable asset, that signals the market sees it as viable. As of August 10, no exchange has indicated plans to do so.

Conversely, if inscription volume drops sharply due to market conditions rather than protocol restrictions, the anti-spam faction loses its primary argument. The dispute becomes academic. Fee data and inscription counts over the next 90 days will determine whether the underlying tension persists or dissipates on its own.

What to watch

Mining activity on the BIP-110 branch. Any sustained block production beyond the current two blocks would indicate the faction has found additional hashrate. Continued silence confirms the chain is dead.

Proof of work change announcements. A formal proposal to switch hashing algorithms on the minority chain would mark the transition from a failed soft fork to an intentional hard fork and new cryptocurrency.

Exchange and wallet support. No major exchange has listed the BIP-110 chain. Any listing would be a significant signal. The absence of listings after 30 days would confirm the market has moved on.

Inscription fee revenue. If Ordinals and Runes continue generating meaningful miner revenue through 2026, the economic argument against data restrictions strengthens. If inscription activity declines, the debate shifts.

Future soft fork proposals. Watch how OP_CTV and other pending proposals adjust their activation strategies in light of BIP-110’s failure. Any proposal that lowers its activation threshold below 95% now carries the precedent of BIP-110’s rejection.

Disclaimer: This article is for informational and educational purposes only. It does not constitute financial, investment, or legal advice. Cryptocurrency markets are volatile and carry significant risk. Always conduct your own research before making investment decisions. Published August 10, 2026.