Mining bitcoin does convert electricity into value though (it's just O(1) as you said it, doesn't scale with more electricity added). Mining bitcoin allows you to exchange value on the network, THAT is what is valuable.
Take regular currencies for instance ($€¥…): they do not convert paper into value, they use paper to represent value. Now does printing more of it makes it better at enabling exchanges? Not necessarily. There's a point beyond which you just get over-inflation (a little inflation can be very beneficial, though).
Now Bitcoin is definitely not at the point where it does causes over-inflation (else its valuation would drop). But you do need to compare it with other currencies, and its energy consumption is a big fat drawback. A deal breaker in my opinion. Barring perhaps some extreme cases, using proof-of-work crypto currencies is simply unethical: too much harm, not enough good.
Mostly, they use mostly-electronic ledger entries to represent value.
Bitcoin needs a reliable way for transactions to work in a single thread (so the block chain doesn't turn into a block tree), and we avoid stuff like double spends. That's proof of work, coupled with a reward (in currency).
To avoid hyper inflation, Bitcoin automatically sets the difficulty so we create one new block every 10 minutes. That limits the number of transactions, but we can remedy that by just making making blocks bigger.
Now the value of a successful block mining is the transactions it represents. Without them, you wouldn't have a means of exchange to begin with. But the reward is a new Bitcoin. And that new Bitcoin by itself does not add any value. It just inflates the currency a little bit, effectively taking value from all other bitcoins. The total value represented by all bitcoins stays the same.
Now we can discuss pros and cons: on the bright side, mining a block enables a bunch of transactions. With a maximum block size of 1MB and 1 block every 10 minutes, we get 7 transactions per second. That's the real value of Bitcoin to humanity: 7 transactions per second, give or take.
As for how much those 7 transactions per second cost, well… there's all the hardware and the resources required to make it, and there's energy. A back of the envelope calculation tells me Bitcoin spent about 8 Gigawatts of power in early 2020, and now it's more like 14. That's 2 Gigajoules per transaction, or about 500KWh.
That's way too much energy for way too little value. Hence my thesis that Bitcoin as a whole is so wasteful that actively participating in it (especially if you're a miner), is unethical.
Worse: It does the inverse of scaling. The more electricity that is added, the less efficient it gets.
Even worse: The system is designed to incentivize more people to join and waste electricity, by paying them to do so.
Moving value is valuable, but the resources expended in moving it do not provide value; rather, they come at a cost, like gasoline in your car.
And Bitcoin is pathetically inefficient at moving value. The average transaction fee has been floating around, what, $15-$30 USD? This is international-wire-transfer pricing range, if it's a single transaction and you don't do these transactions very regularly with your bank.
Don't count the value it provides on the wrong side of the ledger.
Bitcoin is a FULL system, you're seeing the whole thing, the mining is for securing the network AND moving value. Can you say that about any other system (that you are already so used to and blind to how it works)
I deliberately chose international wire transfers as the worst and most expensive way to move money. I assume everyone knows that you can move large sums of money domestically for mere pennies. Yes, that implies the total cost of the system is less-than-pennies per transaction, including human staff, hardware to run everything, taxes paid to governments which protect the system, everything. Yes, transaction costs are the right way to compare the cost of transactions, believe it or not.
Maybe Bitcoin is a FULL system. Big deal. If it's not going to share that FULL system with normal everyday life and its shared institutions, it bears the FULL costs of failing to share, as expensive as that may be.
I'll give you a hint: it's less than 1MWh/tx, which is what bitcoin uses today.
2283MWh per block. 1 MWh per transaction.
Also, you haven't spoken at all about the other side of the equation. Have you accounted for all the externalities for the other monetary network you are comparing Bitcoin to?
Of course not. The existing ones you depend on to protect the necessary infrastructure and servers of bitcoin are already sufficient.
I assume you argue that a single army/government could take out/control miners - which would be the better discussion to have. That's where we would talk about the value of the Bitcoin network.
Centralized exchanges are the reason you don't pay 10000 bitcoin for a pizza. They're the reason why it's even accepter, because there's at least one large place that will exchange your coins for cash.
In my opinion there is little to absolutely no reason to ever change Bitcoin back to cash. Maybe take a loan against or rent it out to others for cash interest, if you really want cash.
I don't value trustlessness, and in fact lack of trust is a vacuum filled by evil.