Recently, the CEO of Ripple stated that more than 50% of the BitCoin miners are in China, therefore, were the Chinese government to interfere, they could control BitCoin (I assume via a 51% attack, I didn't check details).

"“I’ll tell you another story that is underreported, but worth paying attention to. Bitcoin is really controlled by China. There are four miners in China that control over 50% of Bitcoin,”"

Is it true that 4 Chinese miners can control all of BitCoin?

up vote 1 down vote accepted

ViaBTC, a mining pool that favors Unlimited, tweeted “hashpower is law”.

Greg Maxwell (from the Core side) replied “Bitcoin’s security works precisely because hash power is NOT law”.

Introduction

There are lots of people scared of the 51% attack, but in practice, thanks to some changes that have been implanted for years, performing this attack would be harmful to the attacker (in a game theory sense they lose), and not that much to the cryptocurrency user.

How 51% used to be a problem

The main danger that the 51% attack used to face was consistent mining by the 51%. Since everyone takes the longest chain as ground truth, the 51% could mine by itself and not publish any blocks, wait for their chain to be longer, and publish all the blocks, selecting all the transactions inside, and getting all the block rewards.

What if they try

Let's say some team of miners buy all the equipment and accomplishs to have more than 50% of the hashpower in a network, what can they do?

They could in theory perform the attack for a while, and they would get the rewards for mining, effectively fucking over the other miners.

What about the cryptocurrency users? the most damage they could do is to censor transactions for a while, but not consistently, since as soon as one of the other miners can at some point publish a block they could include the transaction.

Can miners change the rules of the game?

No, not even if they have 100% of the hashing power. Client-side validation means that many attempts at mischief that the miner might want to engage in are by default rejected, this is the main difference between blockchains and services like Paypal.

What happens to the attackers

It's important to note that this attack would be fairly easy to detect, not only that, but there are people that base their living on mining, so we would hear about it pretty quickly. As soon as this happens, users could decide to change move their assets, kick off malicious miners, even modify the protocol (the latter one would be the hardest). This would also cause FUD and the price would drop, making the miners, specially the malicious ones to lose money.

Conclusions

A 51% attack takes much more than a 51% of the network to execute nowadays, they barely hurt the end users (delay transactions), it would be easily detected,and it would hurt (economically) malicious miners the most.

Source miners power: https://vitalik.ca/general/2017/05/08/coordination_problems.html

  • Great answer, but could you explain this a little more please? Client-side validation means that many attempts at mischief that the miner might want to engage in are by default rejected, this is the main difference between blockchains and services like Paypal. – VSO Jun 14 at 13:55
  • 1
    Bitcoin has a set of rules, every node checks that the blocks meet those rules, if any node changes the rules and starts behaving differently, they are essentially creating a different blockchain with a different set of rules. For example, business that are running a regular non mining node, or the node that you could be running in your house will reject all their blocks. Therefore if they change the rules they are effectively kicking themselves out of the network. On the other hand, Paypal or Ripple could change the rules and you cannot do anything about it. – Enrique Alcazar Jun 14 at 15:08
  • To be fair, miners could implement themselves a soft-fork unilaterally (This has never happened and it would be an outrage) and restrict the rules, but never to expand or add them. – Enrique Alcazar Jun 14 at 15:13
  • 1
    You mention "kicking off" malicious miners, but without a POW change there is no way to do so. I'm also not sure about your assertion that "A 51% attack takes much more than a 51% of the network to execute nowadays", what is your basis for that claim? – chytrik Jun 14 at 18:38
  • @chytrik There are ways to kick blocks produced from a malicious pool, just by kicking the initial block where they started cheating, all their work goes to waste (look up invalidate block in the API calls). As for "A 51% attack takes much more than a 51% of the network to execute nowadays" You are right I shouldn't have said this without specifying it's my opinion and explaining myself, since I need more space I'll do it in the next comment. – Enrique Alcazar Jun 15 at 8:52

Here is Blockchain.info's chart showing the current estimated distribution of the bitcoin network's hashrate.

Keep in mind: some pools are composed of many individual miners, so in the event of an attack, those individuals may simply redirect their hashpower to a different pool that is not involved in the attack.

In general, controlling the hashrate is not equivalent with 'controlling bitcoin'. If a single entity got control of 51% or more of the hashpower, then they could launch a 51% attack, meaning they could censor transactions. This is an existential threat to the network, but many developers have proposed countermeasures such as a POW change in the event such a thing ever happens.

I would rather say 'dependent' but not 'controlled'. Meaning, China doesn't control Bitcoin on some governmental level, it's just because many miners are based in China and they have a direct impact on Bitcoin's growth. Sure it's a matter of personal opinion, so correct me if I'm wrong.

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Not the answer you're looking for? Browse other questions tagged or ask your own question.