MEV after the switch to PoS

I just watched the new Bankless episode on MEV -> https://www.youtube.com/watch?v=rOVz7dOrGyY (I highly recommend to subscribe to their channel. It really helps to actually understand the Ethereum ecosystem and stay up-to-date on all the new developments)

In the video the Bankless guys interview a panel of three MEV researchers about MEV and the implications of it for decentralized, trustless systems such as Ethereum.

MEV or Maximum Extractable Value is basically all the extra value that a block proposer (either a miner under PoW or a validator under PoS) can extract from the system. Block proposers have this power because they can influence what transactions are included in the block and in what order they are included. For instance, they can "sandwich" (frontrun) trades, they can trigger liquidations and buy up discounted coins, and they can use many other strategies to extract (basically steal) extra value from the network and it's users.

I think the main takeaway is that there does not seem to be a silver bullet to solving MEV, as it appears that MEV is inherent to trustless decentralized systems (you cant just make it illegal like in traditional finance). The best thing you can do is try to guide it in a way that is non-detrimental to the network and make it as transparent and accessible as possible, so everyone is aware of it and miners/validators equally share in the profits so as to avoid centralization due to information asymmetry. Basically this is the goal of these researchers with the Flashbots project.

They also briefly discussed the effects of PoS on MEV and how the current PoS model might actually worsen the MEV problem.

Currently under PoW block proposers are random based on the amount of hashpower they have. So a mining pool with a high percentage of the hashpower knows that it has an equally high chance of being the next block proposer. However it's still random, so they can never be sure. Therefore its very difficult to perform multi-block MEV strategies.

Under PoS block proposers are pre-selected and known at each 12 minute interval. This means that block proposers that are adjacent to each other can not only extract MEV from their own block, but can also collude to perform multi-block MEV strategies.

This has two very serious implications:

1. Multi-block MEV strategies increase the value that can extracted (stolen) from the network and it's users.
2. And even worse, it is VERY centralizing. Basically:
   Higher percentage of validators under your control -> Greater chance of controlling two or more blockproposers that are adjacent to each other -> More MEV that can be extracted.
   This means that centralized staking pools will not only be able to give more consistent payouts (As a solo staker the chance of proposing a block is quite small), but also higher payouts as they generate more MEV. So people will be incentivized to quit solo staking and instead join a centralized service or pool.

Single block MEV seems bad enough, but multi-block MEV could perhaps really harm some protocols. How bad would this actually be? Maybe there are mechanisms that would prevent it? Also would this centralizing effect actually play out in reality or are other existing/planned mechanisms that would counter this? Are there other ways to counter this?

It just seems like a very important topic that deserves more attention from the community. Looking forward to your insights.