I had an interesting debate the other night in Tel Aviv with Charles Hoskinson regarding Proof of Work vs Proof of Stake. Thanks to Ayeka and Zen Protocol for hosting the event.
Charles is a pretty serious guy in the crypto world, having been one of the founders of Ethereum (ETH), the leader of the Ethereum Classic (ETC) split with Ethereum and CEO of IOHK which has created the largest PoS coin by market cap, Cardano (ADA)
Charles's Chief Scientist at IOHK, Professor Aggelos Kiayias, presented an outline of the recent cryptographic proofs regarding the security of Bitcoin's PoW mechanism and the Ouroboros PoW mechanism.
Both these proofs rely on the assumption that an attacker will control only a minority of the hashing (PoW) or staking power (PoS) of the network - hence the risk of a 51% attack. They also rely on the disincentive of a miner or staker to attack or destroy a crypto network which they had invested huge amounts of money in hardware (PoW) or stake (PoS).
I proposed that PoS networks are inherently far more vulnerable to 51% attacks than PoW networks, particularly from the most likely and dangerous 51% attacker of a major cryptocurrency - a nation state.
The reason that PoS networks are so vulnerable to such attacks is that a 51% cryptocurrency stake can easily be purchased in fiat currency which nation states can create at will with no real cost. Thus, unlike other potential attackers, nation states can never have any real stake in a PoS system and thus are not deterred from undermining it by that stake. In addition, because all activity necessary for a successful 51% attack on a PoS network occurs in the online world, it is relatively easy to hide.
In contrast in PoW networks, at least large ones, it is extremely difficult for anyone, including a nation state, to obtain sufficient quantity of the mining hardware required to mount a 51% attack. This is because of real world constraints on supply and a range of other real world issues.
A lively debate ensued where Chris gave me the the moniker "The Miner".
I have further developed my arguments in a more detailed paper on this very important issue for the Crypto Community.