One of the most intense debates in recent years has erupted around BIP-110 — a proposal to temporarily restrict the recording of non-financial data in the Bitcoin blockchain. The initiative, formally aimed at combating spam (images, texts, and other content unrelated to payments), has faced harsh criticism from key industry figures: Michael Saylor, Adam Back, and Samson Mow. The paradox is that none of them are defending spam. Their objections are to the method itself, not the content.
According to BIP-110 monitoring data, miner support for the soft fork has not exceeded 1.5% of new blocks over eight months. It would seem the issue is closed, but the authors designed the activation mechanism so that the miners' position has almost no effect on it: in early August, the restrictions will still take effect, albeit not for the entire network. Let's figure out what exactly the industry found objectionable.
Seven restrictions for one year
BIP-110, also known as the "Reduced Data Temporary Softfork," introduces seven rules for approximately one year. These include a limit on scriptPubKey size (34 bytes, with an exception up to 83 bytes for OP_RETURN), a restriction of individual data elements and witness fields to 256 bytes, a ban on Taproot annexes and unused OP_SUCCESSx opcodes, and a limit on the control block to 257 bytes. Importantly: coins in addresses created before activation are not subject to the new rules. This nuance is often overlooked, even though it removes the main fear — that the soft fork would confiscate existing funds.
However, the problem lies not in the restrictions themselves, but in the method of their activation. BIP-110 uses a modified version of BIP 9, where the miner approval threshold is reduced from 95% to 55%, and the FAILED scenario is absent. If the proposal does not gain support during the voluntary phase, it moves to a mandatory one — nodes that have updated their client will begin rejecting blocks without bit 4 signaling. This is a direct reference to the UASF of 2017, but critics see a dangerous precedent here: a tool that once worked for SegWit could now be used by a minority for a much more controversial purpose.
23 blocks and zero support
The maximum support for BIP-110 among miners occurred in period #475: only 23 blocks, or 1.2%. This is catastrophically far from the required 1109 blocks (55%). Major pools — Foundry, AntPool, F2Pool — did not signal. The only public critic among them was F2Pool co-founder Wang Chun. The situation with nodes is the opposite: the share of Bitcoin Knots, ready to apply the new rules, has grown to 8–20% depending on the calculation method. However, as developer Jameson Lopp notes, these figures can easily be inflated via Tor.
Foundry USA, the largest pool with a share of over 22%, decided to ask its clients. On July 21, it opened a vote where silence counts as "no," and changing the signal requires 51% of the clients' total hashrate. The results have not been disclosed, but with typical turnout, achieving such a percentage is arithmetically difficult.
Crying Luke Dash and a hole found by ChatGPT
In February, Slovak developer Martin Habovštiak recorded a 66-kilobyte TIFF file with an image of a crying Luke Dash Jr., one of the main supporters of BIP-110, into the blockchain. The transaction bypassed all three key vectors of restrictions: no OP_RETURN, no Taproot, no OP_IF. Habovštiak called it a demonstration of the absurdity of the legal arguments of soft fork supporters, who claim the restrictions will protect node operators from illegal content.
An even more serious blow was struck by researcher Dathon Pwn. He fed the BIP-110 activation code to the ChatGPT 5.6 model and asked it to find a bug. The error was confirmed: a node that updates to BIP-110 retroactively does not recheck already accepted blocks. This could lead to chain divergence — not just into two, but into several incompatible versions. A patch has already been published, but the very existence of such a vulnerability in the activation client raises serious questions.
Saylor's 110 objections and the security consortium
Michael Saylor published an essay listing 110 objections to BIP-110. Key points: seven changes cannot be adopted individually — only as a package; the activation threshold is reduced from 95% to 55%, and there is no FAILED state; the rules will be in effect for a year, but the precedent of their adoption will remain forever. "Bitcoin doesn't need guardians of purity. It needs guardians of neutrality," Saylor concluded.
Five days after the essay was published, nine major companies, including Saylor's Strategy, Blockstream, BlackRock, and Coinbase, announced the creation of the Bitcoin Security Consortium. Each participant committed to allocating $15 million over three years for Bitcoin development, but without interfering in the protocol. The consortium's charter draws exactly the same line Saylor wrote about: funding developers, but not taking positions on specific changes.
Who decides
In early August, nodes with the new rules will begin rejecting blocks without bit 4. For the rest of the network, nothing will change. The result will not be a split, but a parallel chain: smaller in both hashrate and number of participants. Eight months of debate have not moved the support signal even to two percent. For the average coin holder, nothing has changed at all during this time: fees, transfers, and balances live their own lives while a war for Bitcoin's future rages on social media.
Cryptalist Expert Opinion: BIP-110 is not so much a technical decision as a political one. It tests how willing the community is to sacrifice network neutrality to combat an annoying but non-threatening phenomenon. The August fork will show how many participants are ready to apply the new rules. But, as Saylor rightly noted, the precedent will remain forever. And that, perhaps, is the main risk.