By allowing you to double spend, it reduces confidence in coins you buy to zero, which reduces the value of all coins to zero. You still have your coins, but they aren't worth anything.
Which is exactly why no one would spend the crazy amounts of money it takes to execute a 51% attack. It would be like self-immolation. Billions of dollars in equipment and energy and you'd have 51% of a worthless network. In practice, you would have to control much more than 51% of the network, because you'd have to catch up to the 49% that are still hashing away.
Unless they are a government, shorted bitcoins, or own a massive stake in a competitor etc.
PS: Remember the value of Bitcoin is limited as a function of the cost of that 51% attack. If the price increases by 10x the transaction fees need to also increase by 10x or Bitcoin becomes less secure.
This idea comes back many times... but I don't see a reason why government would want to short bitcoins or disturb the market to get their hands in it.
1. They have BEP printing press; instead of stealing or brute forcing into bitcoin and then selling the loot for $, the might as well ask Bernanke to print few thousands more sheets of 100 dollar bills (of course not legally but i'm sure there is some overprint like in any business).
2. screwing people out of bitcoins would mean screwing US citizens as well. Why would any part of government do that just to upset Congress and get themselves in front of bunch of congresspeople for grilling? Doesn't make sense.
Government is unrelated to shorting, Governments may chose to destroy Bitcoin for the same reason the may ban it. Basically, currency controls etc.
Someone shorting Bitcoin on the other hand has an economic incentive to destroy it. The cost benefit of doing so scales with the size of their short vs the current hashing power. But, a malware writer may have access to a 51% attack briefly without owning any equipment.
> Which is exactly why no one would spend the crazy amounts of money it takes to execute a 51% attack.
There was a recent paper shared here on an article about new type of currency or exchange system. Although I don't understand details, it explained that since 51% of coins are already mined by just a few pools, if these pools orchestrate together, then can break the chain. But unsure how true this is (cannot find the post anymore, sorry)
Just like why no one would spend $100s of billions on wars that just perpetuate violence.. surely there's no incentive or gain they'd have by causing such disruption.
This is false, you just have to wait for more confirmations. And if miners are censoring transactions there will be a fork where you can use your coins.
Technically, a 51% attack would allow one to confirm non-standard transactions and pretty much do anything. Granted, this would immediately be apparent to any of the other 49% of nodes and cause a fork, but still...
Whats the point of that ? A miner can confirm non-standard transactions right now when he mines a block but no one would accept it. Point of 51% attack is to surreptitiously double spend and nothing else.
You're off by a factor of 20. The current hash rate is 17.5 million terahashes per second. Ant Antminer S9 goes for $2800 and produces 13.5 terahashes per second. In order to get to 51%, you'd need a $2 billion dollar investment.
The US military budget is $600B. Not sure what your point is though. A 51% attack doesn't give a government control, it just allows it to double spend. And that's until a hot patch to increase the required number of confirmations resolves the issue.
Besides there's a huge supply problem here. ASICs need to be manufactured. You can't just buy several hundred thousand of them. The whole point of Bitcoin is that it's prohibitively expensive to just perform a 51% attack and impossible to entirely regulate.
A simple fork invalidating coins originating from the double spend would be enough. You've just spent $2B to achieve something that the network can invalidate with a dozen lines of code. It would set a bad precedent, but it did cost you a lot of money.
If I'm not mistaken, too, the $2Bn investment is just the beginning. The attacker would also have to cover the costs of at least 51% of the energy used in the network - much less find somewhere they can use that much energy without someone noticing.
The idea that a 51% attack on Bitcoin is plausible just isn't sound, imho.
