BIP-110: The Hidden Threat of Bitcoin Loss During Network Splits
Blockchain expert Kevin Loak has identified a critical vulnerability related to the upcoming BIP-110 soft fork. This concerns a real risk of losing funds for those who attempt to sell coins obtained as a result of a network split. This is not just a theoretical threat, but a quite concrete attack scenario that could hit careless holders.
The essence of the problem lies in the double balance that will arise after the blockchain split. The same amount of assets will be reflected in both chains, and this is exactly what attackers could exploit. At an early stage, transactions are valid in both versions of the network, creating a dangerous vector for manipulation.
The mechanics of the attack look as follows: by signing a transfer of fork tokens, the seller inadvertently hands the buyer a ready-made template for an identical operation on the main network. As a result, an equivalent amount is debited from the same address, but in real bitcoins, and fees are deducted twice. At the same time, the wallet is not zeroed out — exactly the volume that was declared for sale is lost.
The most vulnerable will be large holders, who are often the first to react to such events. The only reliable strategy is to keep assets stationary. Copying a transaction requires a signed operation, and if there is none, there is nothing to reproduce. This is an important lesson that was already learned in 2017 after the Bitcoin Cash split, when developers had to build a special mechanism into the new network to prevent the transfer of transactions between chains. The BIP-110 specification lacks such protection.
Why the network ended up on the brink of a split
The conflict has been ongoing since the fall of 2025, when developers released Bitcoin Core v30 and raised the OP_RETURN field limit from 80 to 100,000 bytes. Critics decided that the change would simplify placing extraneous information on the blockchain, so in December the Bitcoin Knots team proposed BIP-110, which limits the volume of non-payment data for a year.
For standard activation, the initiative must be confirmed by 1109 out of 2016 blocks. Currently, about 2.6% of blocks contain the BIP-110 flag. However, the authors also provided another scenario: starting from block #961,632, expected on August 8, nodes with BIP-110 will begin rejecting blocks without the required flag. Since almost the entire hash rate does not set it, these nodes will stop following the main chain.
By itself, this will not create a new blockchain. A split requires miners who will continue mining blocks according to the proposal's rules. If none are found, the alternative chain will not emerge. The BIP-110 restrictions on non-payment data will take effect later — starting from block #965,664, expected in early September. Until that moment, both networks will be able to accept the same operations. To safely separate assets, the owner will first have to obtain coins that exist only in one branch, and only then use them to separate one balance from the other. The speed of block mining could shift both dates by roughly a day in either direction.
My analysis: the situation with BIP-110 resembles a classic dilemma between protocol security and flexibility. Although proponents of the proposal argue that it protects the network from spam, the lack of a mechanism to protect against transaction replay is a serious architectural flaw. Investors should exercise maximum caution and refrain from any operations with fork tokens until the situation fully stabilizes.