Currently, 66% of blocks that get included on the Ethereum blockchain are OFAC-compliant, meaning they exclude transactions from parties sanctioned by the US Government. That's a decrease from its peak and shows that the Ethereum community is serious about keeping Ethereum censorship resistant. 

The 66% figure means that if you have an Ethereum address that is sanctioned by the US government, on average you will have to wait a bit longer before your transaction is included. The good news is - at least from the perspective of people who believe Ethereum should be permissionless and censorship resistant - that this percentage has declined from its 79% peak in the second half of November.

In red the amount of blocks that follow US censorhsip rules

Centralized architecture

But how could this threat of censorship even happen? After all, there is no ‘CEO of Ethereum’ that could find a letter from the government in his mail. The problem started after the Merge, when validators (replacing miners as block producers) started using the MEV Boost service from a company called Flashbots. In this way, they could benefit from MEV. That is the profit that can be made by for example reordering transactions that are waiting in the Ethereum mempool.

The issue is that Flashbots is a near monopolist in the MEV boost space. Even though validators that use their service must also select a third-party ‘relayer’ that delivers pre-built blocks, some of these relayers comply with US sanctions. Including Flashbots’ own relay, the one that most validators simply use by default. Flashbots has chosen to be compliant with the US government. 

But their monopoly is declining. A competitor is bloXroute, which runs three relayers, two of which don’t censor. According to their own numbers bloXroute's relays now provide 20% to 25% of the blocks. 

Flashbots wants to cooperate

Flashbots itself, by the way, is also working on a solution. Interestingly, by incentivizing validators to use their own service less. They proposed a solution called ‘minimum bid’. The idea is to tackle the censorship threat by setting a minimum bid above which validators should use Flashbots. Only the blocks where big MEV profits can be made would be outsourced to Flashbots. The blocks with a small profit, that is, below the minimum bid, would then be built outside Flashbots and thus without being affected by OFAC censorship. 

Conclusion

In the case of Ethereum and other blockchains that allow the complexity that comes with smart contracts, we see a risk of centralization and censorship sensitivity. This emerges from the complicated process of block building in these environments. If only a few sophisticated players can deal with it, a blockchain becomes vulnerable. At the same time, we have seen that the community is aware of this danger and wants to eliminate the bottleneck. Apparently because everyone realizes that ultimately no one benefits from a blockchain that goes against the core decentralization ethos of crypto.