The Futility of a 51 Percent Attack

Bitcoin is more invincible than most people would believe.

Bitcoin is more invincible than most people would first think.

Barring a worldwide lack of interest in either money or the internet, there are exactly three ways to kill Bitcoin, at least in theory. They are:

1.) To destroy all of humanity… Which I doubt needs any explanation.

2.) To use massive Electromagnetic Pulse (EMP) attacks against all civilized pockets of humanity, ensuring no line of communication survives. (A nuke detonated high above every city over 1,000 ppl in size worldwide should do the trick.)

3.) To employ a well timed 51% attack… But I’d like to bring up the argument here that this option really won’t work afterall.

Many people still speculate that government takeover of Internet service providers and of course making bitcoin illegal outright would be able to do fatal damage to Bitcoin too, but these people are simply not paying attention. The former overlooks the invention of mesh networking, which has already been out in the wild for a decade now, and the latter seems to have forgotten that bitcoin is held and accepted all over the planet, not to mention what happened when the US Government made alcohol illegal.

So assuming that bitcoin’s would-be attackers don’t want to die nor live in a world without electricity, the only threat left to discuss is “the dreaded 51% attack.”

The 51% attack for Dummies:

For anyone who isn’t familiar with the term “51% attack,” it is the way that Satoshi Nakamoto described in his original whitepaper on bitcoin to overpower the bitcoin network by controlling 51% or more of the processing power of the entire network. (This includes Miners, wallet holders, all of it. Wherever there is a copy of the blockchain.) Since bitcoin is peer-to-peer, he who controls more than half of the peers could theoretically say which transactions were true and which are false. There are absolutely no other ways to counterfeit bitcoins, because the whole system is built around the concept of majority consensus.

The 51% attack, however, has 3 major flaws, rendering it almost completely useless:

1. The existing bitcoin network is already the most powerful computer network on Earth.

If you take the top 500 supercomputers on the entire planet, (Including every known government supercomputer) hooked them together to calculate the same problem together, then that network would be about 1/8th as powerful at processing the bitcoin blockchain as the current bitcoin mining network is!

Fun fact: That amount only includes Bitcoin miners: Litecoin and other cryptocoin miners aren’t included in those stats. If needed they could jump over to bitcoin and add their hashing power too.

The amount of hashing power fluctuates of course, but is not likely to go down much any day soon, if ever… It’s been growing steadily with the rise of the price of bitcoin, simply because the more a bitcoin is worth, the more miners will invest in mining rigs to mine more bitcoins.

Needless to say, any attempt to make or buy equipment to attempt a 51% attack must happen with bottomless bank accounts. Only two or three governments in the world could even afford to attempt this at all.

2. The Bitcoin community would very easily see the threat coming, and has a “reset button.”

See this chart of the current hashing power of the network? Whenever someone gets a considerable chunk of new hashing power online, such as each time a batch of new ASICs ship, we can all see the corresponding rise on this graph.

Since companies who make the most powerful mining rigs like Butterfly Labs and Avalon ASICs aren’t exactly shy in any way about announcing when & how much hashing power they are releasing to the world, we would easily notice just by looking at this chart, if anyone else came into the network with a lot of hashing power too.

Right now we’re just under one Exaflop in computing power. For someone to attempt a 51% attack on the bitcoin network today, they’d need to bring a whole Exaflop themselves online as miners and once they’ve reached 51%, change the code on all of them at once to favor a new blockchain of their choosing. This would now require far more money to invest in hardware than most countries’ annual GDP.

Even if, however, they managed to bring that Exaflop online in mere minutes flat, (with some magic fairy dust, naturally) and the bitcoin mining community had little to no warning, they’d at least know when it happens and get the word out to the community for everyone to stop mining or whatever so they can investigate.

Since bitcoin itself is a timestamped-ledger and they know the time of the attack, then it shouldn’t be hard for the bitcoin community to “roll back” everyone’s bitcoins to the time just before that attack if most miners agreed to do so.

Indeed, this kind of massive agreement & action already happened recently when the difference between two bitcoin wallet versions hard-forked the blockchain in April, and miners mining with different version of the wallet on their system were able to double-spend for a few moments until it was discovered!

The bitcoin coding community put a temp fix in place within minutes and successfully applied the full versioning fix last week, so this kind of thing can’t happen again in the future. So it’s safe to say that if a 51% attack ever actually succeeded in creating counterfiet bitcoins, there is something the coining community already knows how to do in order to make it like the attack never happened.

3. The incentive to make more money.

Saving the best for last, consider that the number of suspects who would actually benefit from attacking bitcoin is extremely limited… Just 1 to 5 suspects in total, in fact.

Only the heads of the central banks of the top 5 or so nations have any reason to attack bitcoin. There really is no one else on the planet who has the MOTIVE to take down bitcoin with a 51% attack.

Sound like a bold claim to you? Consider the fact that if your goal is to create wealth in any shape, form, or fashion, then you will always be able to make more MINING bitcoin than you would by counterfeiting and of course destroying bitcoin.

In my opinion, this hard little fact is the most genius part about bitcoin, and puts Satoshi Nakamoto up there with Mozart and Einstein in terms of sheer brilliance.

As you add hashing power to the network, you get paid more. To add 100% of the power of the whole network again to the network (required to attempt a 51% attack) means you’d be making as much money as everyone else mining, on the whole planet… This would be no small sum… A constant 25 bitcoins every 20 minutes! At current price, that’s almost a quarter of a million dollars worth of bitcoin every day.

So, if your goal was to make wealth, and you had a spare Exaflop worth of processing power sitting around, would you rather earn a million dollars worth of bitcoin every 4 days, or would you instead attempt to use that power to counterfeit “new” bitcoins, forking the blockchain and at best getting everything shut down?

So it’s quite obvious, there is only one reason to employ a 51% attack: To (attempt to) destroy bitcoin.

Who (of any consequence) would want to do that?

Answer: Only the top central banks who already control a dominant money supply.

So not only would we see a 51% attack coming, and be able to recover from it, but we’d very likely know exactly who is behind it, too.

Would you attack under those circumstances? Considering the cost and the risk?

I don’t think so. Therefore, i’m convinced that bitcoin is safe from any 51% attack.


4 thoughts on “The Futility of a 51 Percent Attack

  1. Pingback: Bitcoin Reaches Critical Mass | Bitcoinfinger

  2. Shmuel Blitz

    I beg to differ. The network’s current hashrate is 85.62 Terahashs/s which is 85,620 Gigahashs/s. A 60 GH/s is typically sold on ebay for about $3,000. for 86,000 GH/s you will need 1434 such devices. That will cost a mere $43,00,000 – well within the range of almost any country in the world.
    Even if the attackers try not to buy from Butterfly labs and, having to build their own ASICs need twice as much devices, the price is still under $100M. Not far-fetched at all!

    1. Luke Post author

      Hi Shmuel,

      The stuff I’m seeing on ebay in the 60 Gh/s range is all backordered stuff like Butterfly labs and Avalon… Might not even ship this year.

      We’re presently at about 98 TH/s, and after the already-sold hardware from the three asic mining companies rolls out, the hashrate could easily be three or more Times that amount.

      The thing is though; the attackers would have to wait in line too right? It will be many months before they start to get their hardware if they went through the normal channels, and then the demand from the other miners would always outpace them for the same resources as long as the price of bitcoin stays high.

      Therefore, they’d need to do their own R&D on ASIC circuitry, create the production facilities, and start creating their own hardware if they really wanted to do a 51% attack from now onwards. These things are not anywhere in the same ballpark as far as costs are concerned.

      1. Shmuel Blitz

        I agree that it will be difficult. But if a big country (e.g. USA, the EU) set their mind to get the needed amount of hash-power, it can (currently) be done with only a slight dent to their financial wings and with not too much of a technologically effort. The numbers might be higher, but still with the reach (financially and technologically) of many many governments.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s