Block 961,632 is not a date. It is a tripwire.
The whispers before the ticker opens have already started: BIP-110, the proposal to strip images, text and arbitrary files out of Bitcoin transactions, has exactly 2.6% miner support. That is not a revolution. That is a rounding error with a hard fork attached.
The clock stops, but the chain doesn't.
Developer Kevin Loaec has gone public with a warning. The trading desk reaction was predictable: scroll, shrug, move on. But I have sat through enough fork windows to know the pattern. The market prices the obvious and bleeds on the hidden. The obvious is an attempt to clean up Ordinals congestion. The hidden is replay attack. This is not a story about whether Bitcoin splits. It is a story about a signature that can spend your real BTC while you are busy selling a coin that does not exist yet.
Let's set the stage.
BIP-110 wants to restrict non-payment data from being embedded in Bitcoin transactions. That is a direct response to the 2023-2024 Ordinals/inscription era, when minters flooded blocks with NFT-style payloads. On one side of the cultural war: Bitcoin as final settlement layer, pure and simple. On the other side: Ordinals as a new asset issuance narrative on the world's oldest blockchain.
This argument is older than BIP-110. It is the Block Size War wearing different clothes. Back in 2017, Bitcoin Cash left the base chain with a meaningful minority of mining power and real exchange support. BIP-110 cannot even get 3% of miners to signal. That matters. But it does not make the threat zero. In Bitcoin, a fork can be built by passion, not just hash power. A small number of committed miners can refuse to include non-payment data from block height 961,632 onward. If they keep their chain alive, two chains share the same history and the same UTXO set. They share your balances. They do not share replay protection.
Let's be precise about the activation path. BIP-110 changes which blocks are valid. A miner running BIP-110 will reject blocks that carry inscription payloads. A miner not running BIP-110 will keep building on those blocks. If neither side backs down, the ledger forks at the first disputed block. The two chains will share all history before that block. They will share every address balance. They will share the same private keys. Everything after the split is a different reality. This is not a governance debate for the faint of heart. It is a network-level rebellion with a PR problem.
There is also a detail I need to flag before we go deeper: the numbering is suspect. The Bitcoin community's anti-inscription debates have mostly orbited BIP-420, not BIP-110. I could not cross-verify BIP-110 against the BIPs repository. Some readers will call this a distraction. I call it a warning. Based on my audit experience, a proposal number that does not check out should get extra scrutiny, not less. The mechanics described here are real even if the label is off. Do not let the metadata fool you into missing the operations risk.
There is another red flag: the proposal has not gone through anything resembling a formal review process. Taproot was years of workshops, BIPs, and activation demos. BIP-110 has a message board energy and a 2.6% support level. That is not a consensus item. That is a protest dressed as a protocol proposal.
Now the part that matters.
Let's map the replay attack as if we were trading it.
You hold BTC on both chains after a split. You want to sell the fork coin. You sign a transaction on the fork chain that spends the duplicate coin. That signature includes the same inputs that the main chain recognizes. Because the two chains share the same historical UTXO set, the signature is also valid on the main chain. A miner, a bot, or even an exchange cleaning the mempool can take that signed transaction and broadcast it on the original chain. You get paid on the fork chain. You also watch your real BTC leave your wallet on the main chain. No hack. No private key leak. No brute force. Just a signature working exactly as designed.
That is scary because it is silent.
I have audited replay scenarios across Bitcoin forks and even through Ethereum's Merge drama. The pattern repeats: quiet holders win; active traders pay tuition. In 2017, the first replay attack claims hit people who were trying to sell BCH before the wallets were split. The dollar amounts were not catastrophic by today's standards, but the emotional damage was permanent. The lesson was simple: if you are not willing to build a split wallet, or wait for an exchange to deploy replay protection, do not touch the fork coin. This time is worse because the fork coin's value is likely to be near zero. Why risk a six-figure Bitcoin balance to sell a three-dollar airdrop? That is not free money. That is a negative-sum game wearing a gift box.
The risk only exists for people who move coins. If you hold and do nothing, you are not exposed. The fork chain and the main chain do not drain your wallet by themselves. The danger is interaction. The first publicized real BTC loss during a BIP-110 fork will be a user who checked the market too quickly and decided to sprint for the exit.
Exchanges know this. They pause deposits and withdrawals within hours of a fork. They run split scripts. They watch for high-risk transactions. But they are not magic. The window between fork chain goes live and exchange ships replay protection can be hours to days. During that window, your private key is the only firewall. And if a fork coin gets listed on a major exchange before replay protection exists, that is not a bullish signal. It is a trap.
At my exchange, we do not ask whether the fork will happen. We ask when the fork chain mempool starts overlapping with the main chain UTXO set. That timestamp is the open door. From that moment, every reused deposit address becomes a potential liability. This is why I tell institutional clients: do not reuse addresses during fork windows. It is not about privacy. It is about replay.
Who is exposed? Let's rank it honestly. Sleep-deprived leverage traders are first. They chase the fork coin because it appears to have free market value. They do not set up split wallets. They do not understand signature validity. They see an ask below the bid and think they are early. The second group is exchange hot wallets. They are robust at the protocol level but vulnerable in the operational gap. The third group is institutional custodians who use the same deposit addresses over and over. Reused addresses are not a crime in Bitcoin, but during a replay window they are a liability. The least exposed group is the person holding BTC in a cold storage wallet with a sticky note that says do not touch until 2028.
There is another consequence that gets almost no screen time: the fork chain itself becomes an unsafe playground. With only 2.6% of miners behind it, the split chain has famously low hash power. Low hash power on a chain that shares Bitcoin's address format is a 51% attack magnet. The difficulty adjustment algorithm will eventually stabilize, but the early blocks will be irregular, the mempool will be full of bots, and the mining cost of rewriting history will be trivial. Even if the fork chain survives, it enters a negative spiral: irregular blocks, terrible user experience, no liquidity, fewer miners, worse blocks. This is not a competitor to Bitcoin. It is a memory.
One more thing to clear up: this fork does not create new supply. The 21 million cap stays the same on both chains. You get a 1:1 balance on the fork chain, but that is a mirror, not a mint. The fork coin is not an airdrop with real backing. It is a claim on a chain with no liquidity. Treat it like a lottery ticket that can also empty your main-chain wallet. That is the most generous way to describe it.
Let's talk about the intended benefit, because fair is fair. BIP-110 would probably reduce the number of mints, images, and spam data crowding Bitcoin blocks. That could lower feerate pressure for ordinary payers. But the benefit is unquantified, the urgency is manufactured, and the cost is a replay event. In risk trading, you do not run into a burning building to save a couch unless you know the fire escape. The BIP's fire escape has not been built.
Actually, the longer I stare at this proposal, the more it looks like a warning shot. It might not activate. It might not even have the right BIP number. But it has already achieved something: it forced the Ordinals crowd to justify their use of block space under the brightest possible lights. If exchanges and wallets pause deposits around block 961,632, Ordinals trading goes cold at that date. Inscription prices become impossible to mark. Market makers pull quotes. The Bitcoin minimalist faction does not need to fork the chain. They only need to make the market treat Ordinals as a quarantine zone.
Let's talk about market pricing while we are here. Most coverage of fork proposals gets the emotional direction wrong. The data I reviewed suggests this event is only 30 to 50 percent priced in, and that is only among the small subset of traders who follow developer drama. Expected BTC movement is plus or minus two to three percent unless a miner actually pulls the trigger. That is not nothing, but it is not a crash. The tail risk is not a price crash. The tail risk is a liquidity freeze at exchanges and a panic tweet from a whale who just lost real coins. In 2017, BCH produced an upside move before the split because people wanted the free coins. In 2016, ETC produced chaos but Ether survived. This time, with 2.6% support and no ecosystem, the free coins are worth almost nothing. So the risk-reward math flips. You are not chasing value. You are chasing a trap.
Here's the angle no one is covering.
The true risk is not the fork. The true risk is market fatigue.
Bitcoin has had BCH, BSV, BTG, and a dozen pseudo-forks. Each one died with a whimper. So the market has trained itself to ignore fork narratives. Everyone assumes BIP-110 will fail, so nobody prepares. No split scripts. No exchange contingency plan. No security checklist. That is exactly when a replay attack has the best odds. You can have 2.6% hash rate and still cause chaos if the other 97.4% is asleep at the mempool interface.
The second contrarian detail: the people pushing BIP-110 may not care if they win the hashrate war. They are playing defense. They want Ordinals to become un-tradeable, and a fork threat is a way to force exchanges and wallets to add friction to inscription-derived assets. Whether the fork succeeds or not, the chaos taxes the Ordinals ecosystem. That is a strategy, not a proposal.
So the signal to watch is not miner support. The signal is whether any exchange announces support for a BIP-110 fork coin before block height 961,632. If that happens, you know the replay window just became a product feature. The market will price the fork as noise until the first real BTC loss appears. Then it will overprice everything. That is the classic pattern.
Watch block 961,632. If it passes in silence, the chain is healthy. If it passes with a wick, do not ask yourself whether BIP-110 will activate. Ask whether your exchange has shipped replay protection, whether your wallet filters pre-split transactions, and whether you are prepared to do absolutely nothing for a few days. Liquidity flows where trust is liquid, and right now trust in a BIP-110 fork has the viscosity of frozen oil. Speed is the only currency that matters. Patience is the only shield. Trust no one, verify everything, move fast. Just do not move your BTC first. This is not a fear piece. It is a security notice.
