A Comparison Of LN and Drivechain Security In The Presence Of 51% Attackers

Summary:

In a recent post on the bitcoin-dev mailing list, developer Paul Sztorc responded to claims that recursive covenants are a bad thing. Sztorc acknowledged that it is easy for a 51% miner to double-spend in the Lightning Network (LN) by censoring "justice transactions", but he also pointed out that a 51% miner can only attack LN channels that it is a participant in. In contrast, a 51% miner can simultaneously attack all Drivechain-based sidechains and steal all of their funds. Additionally, users of the LN have some protection against 51% miners because there is a probability that they make a channel with a non-51% miner. Sztorc explained that if Alice wants to attack Bob, she may try to convince the 51% miner to help her steal from Bob. However, the 51% miner would prefer receiving payment now from Bob over the promise of payment from Alice later. Moreover, if Alice offers to pay the 51% miner now from some onchain funds she has, she risks losing her channel amount and having to wait out the `OP_CSV` delay to get the same amount later. The honest participant always has the upper hand, even in a 51% miner scenario, due to the `OP_CSV` delay. Sztorc also noted that with Taproot, a mutual LN channel close is indistinguishable from a singlesig spend, so not all LN-related transactions can be censored by the 51% miner. However, in the presence of channel factories, the entire factory is at risk if one participant is the 51% miner or a sockpuppet thereof. Thus, channel factories trade off further scaling for reduced protection against 51% miners.

Updated on: 2023-06-03T07:41:25.440924+00:00