I read Why Bitcoin Matters. Andreesson description of the Byzantine General's problem is superb., but his shitcoin thesis is flawed. For example, @Tim Bouma made an online locker using digital signatures... So Andreeson is not totally wrong about describing this possibility in 2014, but these lockers do not require a blockchain. The Byzantine general's problem does not apply to to information that does not require a public stamp(timestamp) or metal coins(year of minting). The purpose of these stamps is to audit the supply of money(or precious metal) at some interval. With metal coins, the metal is audited once. As time passes, the quantify of metal decreases because of wear & tear and debasement. Bitcoin verifies every coin on the network, not once, but every ten minutes on average. In the analog world, you can only put one date on a coin. That date must last forever and it takes a lot of equipment and expertise to assay the metal in an old coin. Bitcoin "assays" the math of each sat 52,596 times a year, each and consensus is reached by every participant in the network. That's the technology leap of bitcoin. It's a coin, not a contract. Satoshi Nakamoto wrote, "Digital signatures are part of the solution, but the main benefits are lost if a trusted third party is still required to prevent double spending. We propose a solution to the double spending problem?" AI does mean smart contracts could now replace a repo-man. If the buyer fails to pay the car payment, the car can drive itself back to the dealership. But guess what, that doesn't need a timestamp on a blockchain. The bank just needs to prove the customer is a deadbeat. If I send my car payment to the dealership every month, there's no need for my car to drive back to the dealership. It's a dumb contract, it a smart contract. We don't need another blockchain to tell the dealership If I don't pay my car payment. They'll know even though it's not on a public blockchain. They can have their agents summon my car back to the dealership without a shitcoin.