Given the rapidly rising difficulty level and electricity consumption, is the protocol sustainable as-is, and if not what is most likely to happen in the evolution of BTC?
As Bitcoin expands there will be much greater numbers of transactions while the cost of maintaining the network will remain relatively constant, if not decrease, as power-inefficient GPUs and higher consumption ASICs are replaced by more efficient ASICs. The cost per transaction will thus drop.
Also, keep in mind that it isn't replacing nothing, it's replacing centralized accounting which has a significant cost per transaction.
Reducing the number of miners will reduce the cost/transaction. As the reward decreases the number of miners will also decrease. In the near term (2-4 years) ASIC increases in speed and reduction in power will stabilize. As mining becomes less profitable there will be fewer people mining thus stabilizing the transaction cost.
One needs to look at BTC from the standpoint of a currency, not from the standpoint of a miner--looking to get rich mining.
Yeah, I mine.
As of 2023 the energy usage is already mother huge: > 0.3% of world electricity production. By design (proof of work). This can become even deadlier problem if it's usage picks up even more, and the price goes way up.
The problem is that proof of work distributes the fixed block reward (which is huge - currently about 9 billion $ per year at bitcoin price $20000) randomly among miners according to their hash rate, setting an incentive for them to increase the hash rate until the block reward barely covers the costs, which are mostly electricity. So the global energy cost spent on "mining" tends to be a major part of block reward, which is proportional to its price up to the halvings, and that's little dependent on technological advances in mining or otherwise.
I don't think it's possible to make an actual prediction what happens because that depends on many external factors that influence the demand for bitcoin and thus it's price. But the problem I described has the consequence that as long as Bitcoin is using proof of work as it does right now, it's price cannot go up indefinitely. So there are basically two scenarios:
the price stays low enough so that the mining reward doesn't get much higher than it currently is or at least not long enough for the mining to adapt to that price. Note that admits a slow raise because of the halvings. Or:
if it doesn't stay low (some predict the price goes up to > $100000, e.g. because commercial usage goes up), the mining resource usage will be that high that politics has no choice but to intervene, e.g. by pushing it's price down again by outlawing it's commercial usage until proof of work is exchanged to something else, like Ethereum did. (This is pure speculation, but if you consider the effect of rumours about ETFs and the SEC attacks on bitcoin price in 2023, you can imagine what a full blown ban on commercial bitcoin usage by the EU and the US would do, possibly enforced by sanctions. In fact the EU already considered something like that, but did not (yet?) do it.)
Just for the fun of future readers: my personal opinion is that it'll go up, politics will intervene in two years, and there will be a modification of POW that will be pushed as the "official" Bitcoin and the unchanged POW-Bitcoin will live on under the radar at low prices. But that's the part where I just guess.
If you like even more details than it's possible to give here - I wrote a blog that contains this topic.
Update: it turns out you can easily verify that this matches the current numbers: the current mining consumption is more than 50% of the block reward, and the block reward is also the upper bound for the money spent on electricity as it pays for the mining.
It appears to me that electricity costs to mine blocks are rising at an unsustainable rate. Charts of difficulty and network hashes are growing exponentially. This is unsustainable. Currently miners generate revenue of approx. $30 on every transaction. Who pays for this? BTC holders, that's who. (Source: Blockchain.info charts) While it appears that fees we pay to transact are low, we actually pay this much higher amount mainly through inflation right now, due to mining of new BTC.
The big claim in favor of protocol sustainability is that the market will find a balance at the right transaction fee level. I would argue that already it is being shown that effective transaction fees (mining inflation + fees) are too high - mainly because of the underlying costs of producing blocks, of which electricity is a growing component. An unsustainably growing component. And that the low fees we pay to transact are actually an illusion hiding the real costs simply because so much investment money is flowing into the illiquid BTC base, preventing the inflation effect that would otherwise reduce the BTC price being so apparent.