> LN doesn't really solve the energy problem because the incentive for miners to compete is still dominated by the mining reward rather than transaction costs.

You're right, it doesn't solve the energy problem. LN solves the fee problem. The energy problem is addressed by something like proof of stake.

> Proof of stake is interesting but I don't think it's been used at scale yet has it?

It's been used in a number of non-major cryptocurrencies. Its been analyzed quite a bit though, and there are high-value (in the sense of market cap - i.e. large bounty for someone to exploit it, if it were possible) working implementations out there.

You haven't solved the fee problem if you haven't solved the energy problem.

I'm not sure what your reasoning here is, but you're wrong. LN basically allows you to dramatically expand the number of transactions per block. The blocks still take energy to mine, but the per-transaction fee gets sharply reduced.

Do you see a solution to the energy use problem that doesn't involve a massive devaluation of Bitcoin?

Right now, proof of stake seems like the best solution to the energy problem. AFAIK, bitcoin is not looking to implement PoS, so my bet is on Ethereum to address this issue.

There is no “energy problem”. Energy spent mining bitcoin is the price paid for security of the ledger in in-cheatable currency of the universe. Proof of stake systems are all castles in the air in that regard.

If they are castles in the air, then why do they work?

markets are often irrational.

Ok, but proof of stake is a security model. So what you're claiming is not just that markets are irrational, but that thieves can steal all this money, and are just choosing not to. Since people are constantly stealing money in the crypto world, I find that pretty hard to believe.

in proof of work security is paid for in energy. in proof of stake security is paid for in currency of the system. if this circular property doesn't bother you - fine, but it bothers me.