generally, crypto just won't matter very much going forward. some chains * stablecoins will likely get significantly more use and tx volumes as autonomous agents become real. but this will be an implementation detail, the real societal changes will come from the agents and their impact on labor, not from their transaction layers or whatever byzantine ownership structures people build.

Ethereum is useful, but so is Stripe, and it's tough to see why ETH would deserve a multitrillion dollar market cap even in a world where it was very frequently used financial infrastructure.

people talk about the future of crypto past basic financial primitives as helping solve coordination problems. and it does do this in some very cool ways, eg Dai is a marvel of incentive engineering. but past the financials, i think coordination problems will get *easier* rather than harder with the intelligence explosion. you can incentivize agents to achieve goals without paying them. that's kinda the point.

long term, bitcoin as the protocol currently stands will not survive.

price fluctuations aside, the narrative has been that the current system, the current clients, provide an immutable perpetual secure store of value. objectively this is not the case. the security budget is programmed to run out.

the response there has always been that transaction fees will make up the difference and incentive miners. this is empirically not the case. in 2025 fees made up about 1% of miner revenue. there is no structural reason for fees to increase along with Bitcoin's market cap. especially with the dominant "store of value" narrative, its biggest users and holders and advocates are definitionally not frequent transactors.

"the price will keep going up!" no, you're not listening. sure maybe, but the point is that there isn't actually a corresponding increase in the amount paid to miners. transaction fees might be denominated in BTC but the cost is just a free market, if there's not a demand for transactions the fees don't go up. the price increasing is actually a problem, as it *reduces* the percent of the market cap that the security budget makes up.

block rewards are still relatively high now, but in 20-30 years your immutable perpetual asset is increasingly cheap to attack.

and actually this is much less of a long-term issue than you think. even with "high" block rewards, nowadays current miners have massive opportunity cost for their facilities vs just becoming data centers. bitfarm, mara, riot, etc, miners are already pivoting en masse.

so what's left? large holders eat the costs of mining themselves to keep their assets secure? not a great incentive structure, just makes everything dependent on a few large player's continued investment. some kind of side chain massively increases transaction demand and fees? it won't be lightning, that's just not structurally meant to land to the main chain that often.

the protocol has to change to survive. so shut the fuck up about "perpetual stores of value" and figure out how to actually make it functional.