[everfree]

The writer of the article supposes that it's a critical vulnerability that a validator can come up with an "equally valid" alternate history for Ethereum that stems from a fork several months prior, after the validator has already withdrawn their stake from the canonical chain to prevent slashing.

In reality, everyone just ignores the attacker's chain, because it attempts to rewrite not seconds, minutes, or days of canonical history (periods of time short enough that the identity of the canonical chain might be in question), but many months of history that the entire world has already agreed on and that node software refuses to reorganize the chain past.

This post by Vitalik in 2014, shortly before the launch of Ethereum, explains why this "weak subjectivity" property of PoS is not nearly as problematic as PoW proponents make it out to be.

https://blog.ethereum.org/2014/11/25/proof-stake-learned-lov...