Perhaps China could decide that Satoshi's blocks should be given away to charity.
Aha, sure. Good luck going back to the first 210,000 blocks, and reversing years of history. And not only that: good luck rejecting existing soft-forks, which were activated at certain block heights, and writing a chain, where those checkpoints will be in a different block numbers, or with different block hashes, without alerting thousands of nodes.
More than that: good luck reorging thousands of blocks, from more than one halving, and triggering all block explorers, and other databases. And also good luck triggering all pruned nodes, that they have to re-download the whole chain from scratch, and rebuild their database. Good luck doing all of that unnoticed, without a single topic on the forum saying "ALERT! HUGE NETWORK REORG!".
Is this technically possible?
Technically? Maybe. But if you ever tried to mine 100k blocks or more, even in regtest, with minimal difficulty, on CPU, then you know, how slow it is on CPU, when you can mine hundreds of blocks per second. And imagine, that you don't have regtest difficulty, but mainnet difficulty. Even using "the longest chain" instead of "the heaviest chain" means, that your regtest experiment is 2^32 times easier to execute. But if you take into account the current chainwork, then guess what: collecting one million newly-created BTC in fees alone is much, much more profitable, than executing such attack.
To sum up: mining new coins on charity's address is easier and more profitable, than forcing exactly Satoshi's coins to land there.
What would be the technical implications of China doing this?
1. Rejecting existing soft-forks, and re-activating them under new rules (or leaving them unactivated).
2. Rebuilding databases in a lot of nodes in the world, triggering a huge debug log report for all node runners.
3. Triggering blockchain redownload for many, if not all, pruned nodes.
4. Alerting a lot of full node runners, that they are back in "Initial Blockchain Download" stage, which will cause a lot of questions.
5. Deciding to increase the supply or not, by deciding, if some coins should be burned in the coinbase transaction, in OP_RETURN outputs, and so on.
6. Seriously changing the total chainwork, so it will produce even more detailed debug logs in all node runners.
7. Being blacklisted by a lot of nodes, because of exceeding P2P network traffic limits, and being marked as "misbehaved node" by most of the network.
And so on, and so on. If you want to experience it, then use your CPU, run Bitcoin Core in regtest mode, mine 800k blocks, and try to trigger for example 600k block reorg. Good luck. If you see, what happens, when you do it on your CPU, with minimal possible difficulty, then you can go further, if you are patient enough to observe it.
Once I mined something like 250k blocks in regtest. It took way too much time, to not be noticed, even if it would be some kind of public test network.
Edit: One more thing to note. Currently, developers are trying to kill testnet3. It is just a test network. It was resetted previously two times, and it is going to be resetted again. You can see, how many people are complaining about it. You can also see, how many people are complaining, where some altcoins are resetted, hard-forked, or changed without reaching consensus. Judging by that data, you can only imagine, how loud people could scream, if similar things would happen to the mainnet.
Even with fiat currencies, if you mess up with money, you can make a war. In the crypto world, people are even more focused on some values like decentralization, and all newspapers would be flooded by headlines like "BITCOIN HUGE CHAIN REORG" if you would try to even trigger more than 100 confirmations chain reorganization, not to mention more, like reorging beyond more than one halving, to get access to Satoshi's coins, and to not let them to be sent to Satoshi in the first place.
