Can Bitcoin withstand attacks by state-level agents?

[santandar]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

[DesmondHayes]

Hello new fellow, imagine if someone with a lot of super-powered computers ( state - level agents) decided they wanted to mess with Bitcoin. They could try to change the history of Bitcoin transactions and maybe even take some coins that aren't theirs. This would cause a big mess and people might stop trusting Bitcoin. But don't worry! The people who take care of Bitcoin are really smart and would work together to stop the bad guys from doing that. Plus, Bitcoin is designed to be really strong, like a superhero, so it's not easy for anyone, even a big government, to mess with it too much. So, while it's something to keep an eye on, we can feel pretty safe knowing Bitcoin has some tough defenses!

[NeuroticFish]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Can you understand that in order to do 51% attack they would have to either buy some of the pools to redirect their power, either buy as many mining devices as needed to match the current power (let's approximate to buying as many mining devices as they are in use now)? Or, well, a mix of the two. I myself I'm unsure of the real magnitude of this number (of devices).

The thing is that even big mining companies may have to stay in the waiting list to buy the gear they need. So it would be a very slow and expensive task, not mentioning that since those billions would be taxpayers' money, if the government is a democracy, it will be the last thing (use of money) they'll approve before getting changed and the preorders canceled.

[Amphenomenon]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Can you understand that in order to do 51% attack they would have to either buy some of the pools to redirect their power, either buy as many mining devices as needed to match the current power (let's approximate to buying as many mining devices as they are in use now)? Or, well, a mix of the two. I myself I'm unsure of the real magnitude of this number (of devices).

In the end 51% is not possible, Bitcoin is different from every other POW blockchain especially since the equipment to run this is infeasible

The thing is that even big mining companies may have to stay in the waiting list to buy the gear they need. So it would be a very slow and expensive task, not mentioning that since those billions would be taxpayers' money, if the government is a democracy, it will be the last thing (use of money) they'll approve before getting changed and the preorders canceled.

Even if most government always speak negatively about bitcoin they can't deny the fact that they need bitcoin to grow aside from the taxes some are also investors in a nutshell government will actually protect bitcoin from this.

[pooya87]

It depends on the type of the attack. Not everything is 51% attack nor is it the worse thing they can do.

There are a lot of principles in Bitcoin they can attack and cause as much damage to Bitcoin's reputation. For example censorship which is one of the main principles of Bitcoin (being censorship resistant). This has been under attack for some time now. For example in United States which has the biggest anti-privacy regime, they have been trying to force censorship in Bitcoin world from all the restrictive laws imposed on businesses (like centralized exchanges enforcing terrible KYC laws) to forcing miners and pools in their jurisdiction to literally censor transactions (like the case with MARA pool).

This is not making as much noise as threat of 51% attack because it is not as fancy an attack as 51% attack and it is not yet happening on a large scale.

[Wind_FURY]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Can you understand that in order to do 51% attack they would have to either buy some of the pools to redirect their power, either buy as many mining devices as needed to match the current power (let's approximate to buying as many mining devices as they are in use now)? Or, well, a mix of the two. I myself I'm unsure of the real magnitude of this number (of devices).

The thing is that even big mining companies may have to stay in the waiting list to buy the gear they need. So it would be a very slow and expensive task, not mentioning that since those billions would be taxpayers' money, if the government is a democracy, it will be the last thing (use of money) they'll approve before getting changed and the preorders canceled.

Plus even if they were successful in buying all the mining hardware required, and actually did a 51% attack, the community, the economic majority, and its army of full nodes would make an emergency hard fork and kick the attacker out of the network. The state-level actor spend billions for a mere double-spend.

But we could be very VERY confident that during the time the state-level actor was accumulating the hardware for its attack, it will realize that what's keeping the network together is its incentive structure, and that it's actually better to be an honest actor in the network and be incentivized in Bitcoin.

[DaveF]

Computationally, not easily or simply or cheaply.
Other ways, very easily. There have been bans and other things with crypto over the years, but nothing ever harshly enforced.
If tomorrow morning some government passed an outright ban and long jail times for using, trading, having or doing anything with crypto it would be a lot more effective then trying to shut it down another way.
If other countries / governments joined that cause then it could happen.

But if they did that, then they could not tax it. And no government wants to give up tax money.

-Dave

[seoincorporation]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

You have the wrong concept about the 51% attack, if a person or a company have 51% of the network they could make millions of dollars, let's say they send 10,000 BTC to an exchange, and sell them, once they get the money then use the 51% attack to double spend the 10k btc transaction for the exchange, that way they recover the bitcoins that they already sold. That's the real danger of this attack.

If a state wants to force changes in the bitcoin protocol, then the community will not allow it, the community can always fork the code and fix any forced change made in the past.

[SickDayIn]

There are two different attack chain scenarios that advanced persistent threats, or nation state actors could attempt:


1) A 51% attack targeted at controlling 51% or more of the total mining power to fork Bitcoin and create a new primary Bitcoin chain.

2) A quantum computing attack on the cryptography supporting Bitcoin itself.

Whilst both are attack scenarios which can be considered, I think your thread focuses the point of discussion on the 51% attack.

Now, the way to conduct this attack would either require purchasing 51% of the existing compute power from miners (e.g. AntPool, etc), or purchasing new compute power, at 100% of the current available compute used by miners, thereby doubling the available compute of the current 100% to 200%. This is probably more likely as it would make no sense for miners to sell their hash power to a state actor who had the intention of destroying Bitcoin. Realistically they would only be successful an at attack by introducing new compute and doubling the overall supply.

However even then whilst 51% could be required theoretically, forking Bitcoin would require much more compute, like 60-70% or more. Just look at Bitcoin Cash, Bitcoin SV, and the other forks that exist as examples why.

Edit: Update to this post, I have been corrected by @ABCbits, you can actually perform this type of attack without a 51% hashrate successfully, but on a single transaction with only a few confirmation, a good calculator is here: https://jlopp.github.io/bitcoin-confirmation-risk-calculator/

[Kakmakr]

The worst case scenario for those countries are that they are spending Billions of Dollars on attacking Bitcoin and once they do that, then people simply switch to some other Alt coin and they wasted all that tax payer money on a fruitless attempt to take over a technology that can be replaced by  another Alt coin with the same features.  

You will also have to sustain that attack, so you will constantly have to throw loads of money on that attack to basically do some double spends until this is noticed.   ( There are a lot of eyes, monitoring the Blockchain)

The reputational damage for Crypto currencies  will have the biggest impact, if this succeeds. ( People will lose trust in Crypto currencies in general)

[ABCbits]

Now, the way to conduct this attack would either require purchasing 51% of the existing compute power from miners (e.g. AntPool, etc), or purchasing new compute power, at 100% of the current available compute used by miners, thereby doubling the available compute of the current 100% to 200%. This is probably more likely as it would make no sense for miners to sell their hash power to a state actor who had the intention of destroying Bitcoin. Realistically they would only be successful an at attack by introducing new compute and doubling the overall supply.

I don't think there are any mining pool which offer to sell hashrate from it's miner. While there are marketplace (such as NiceHash) where you can buy/sell hashrate, the amount obviously not enough to perform 51% attack on Bitcoin. And buying new ASIC to match current total mining hashrate isn't realistic either. Aside from the cost, there's limited amount of ASIC they could buy within short time.

However even then whilst 51% could be required theoretically, forking Bitcoin would require much more compute, like 60-70% or more. Just look at Bitcoin Cash, Bitcoin SV, and the other forks that exist as examples why.

Wrong, theoretically total mining hashrate doesn't matter. The forked coin can run with far smaller hashrate. And if that forked coin switch to PoS (or other similar algorithm), mining hashrate isn't even needed.

[BlackHatCoiner]

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Here's a list of what I predict will happen (with no particular order):

• Billions of tax payer's dollars would be wasted. The Bitcoin network would continue operating, as the state cannot endlessly retain this attack, because it doesn't have the money.
• People would likely consider Bitcoin transactions prone to reversal for quite a lot of time.
• If this happens in the far future, then perhaps the government that attempted to do it will be sanctioned by other governments that save / invest in Bitcoin.
• Some people will take advantage of their transactions being unconfirmed and double-spend them.

Here's a list of the reasons I think this is unlikely to ever happen:

• It'd require an astonishingly efficient coordination between the government people. And usually, they don't work efficiently, because... they're bureaucrats.
• It'd require vast amounts of money such that they'd be more incentivized to issue and earn the number #1 asset in the world, than destroy it.
• They would have to purchase an unusually large number of ASICs, and setup a noticeably long infrastructure. I don't believe there even exist so many ASICs, so they'd probably have to manufacture their own.

Did I mention that bureaucrats do not work efficiently?

[cryptosize]

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Here's a list of what I predict will happen (with no particular order):

• Billions of tax payer's dollars would be wasted. The Bitcoin network would continue operating, as the state cannot endlessly retain this attack, because it doesn't have the money.
• People would likely consider Bitcoin transactions prone to reversal for quite a lot of time.
• If this happens in the far future, then perhaps the government that attempted to do it will be sanctioned by other governments that save / invest in Bitcoin.
• Some people will take advantage of their transactions being unconfirmed and double-spend them.

Here's a list of the reasons I think this is unlikely to ever happen:

• It'd require an astonishingly efficient coordination between the government people. And usually, they don't work efficiently, because... they're bureaucrats.
• It'd require vast amounts of money such that they'd be more incentivized to issue and earn the number #1 asset in the world, than destroy it.
• They would have to purchase an unusually large number of ASICs, and setup a noticeably long infrastructure. I don't believe there even exist so many ASICs, so they'd probably have to manufacture their own.

Did I mention that bureaucrats do not work efficiently?

They'd have to spend insane amounts of electricity (inb4 energy shortages) to... prove their point. Would Greta approve that?

[d5000]

I recently wrote something about this case in another thread.

First, yes, such an attack is of course possible. Besides of the high cost even for states, there is however also another problem. If the state-level agent wants really to harm Bitcoin, he would need additional measures which would make the attack even more expensive:

First, he should be able to maintain the attack for some time (was already mentioned). But second, to really initiate a big dump, he should create real damage, e.g. double spending several big amounts on big exchanges or payment processors.

For this to happen, he'd need to create a network of fake accounts on several exchanges, bypassing KYC measures. Of course he has to take into account that these exchanges have some safety measures in place, such as requiring confirmations for BTC deposits. It is likely that the attack would be noticed already in the timeframe between the 51% incited re-org and the moment the deposit gets "credited", or at most the intent to cash out by the attacker, and the exchange would simply refuse to credit BTC deposits which were the result of a double spend.

Atomic swap and no-KYC exchange liquidity should be too low to create such an "real damage" attack.

The state-level attacker could of course get, in theory, some of his costs back if he manages to short Bitcoins just before the attack. But shorting currently is only possible on centralized platforms. If these detected anomalies like a considerable number of shorts just before the attack, they would for sure initiate investigations. And the state-level agent would then be responsible for any damage created to nationals of other countries by the "short selling" part of the attack, because "insider trading" is a crime in all countries I'm aware of. So no, the most likely outcome is that he would get next to no "returns" from the attack.

What they could try is to inflict damage with the "cheapest possible" 51% attack, i.e. with only a small amount of double spends and only for a couple of blocks. But the harm created would then also be quite low. Several coins which were attacked in the past, like ETC, have since then recovered.

I think thus for state-level agents the "FUD attack" is much more promising: tighten regulations, trying to create narratives like "Bitcoin is harmful for the environment", or outright ban Bitcoin, the latter being probably possible only in authoritarian countries.

[WatChe]

And buying new ASIC to match current total mining hashrate isn't realistic either.

To acquire powerful advanced application-specific integrated circuit (ASIC) miner like Bitmain S19 XP Hydro with hash rate of 257 TH/s will costs around $6,000 (new). One need scores od such devices to carry out 51% attack on bitcoin. And not to forget the electricity to power these machines, it will be difficult to find a source that can power all these machines at a time. 51% attack exists on paper but with Bitcoin its something not possible practically.     

[BlackHatCoiner]

They'd have to spend insane amounts of electricity (inb4 energy shortages) to... prove their point. Would Greta approve that?

Of course she would. Just pay an insane amount of electricity once to ruin the currency that "destroys the environment".  

[...]

This explains well why state-level agents wouldn't attack the network with the intention to steal bitcoin, for example by purchasing billions worth of bitcoin and then shorting them right before they reverse the transactions. Doing so would require a lot of time, which is translated in a lot of blocks, and therefore a great cost in reorging those blocks.

But, in my opinion, it makes no sense for the state to attack like that. Instead, it's more likely that one of the following will happen:

• It already has an effective way to steal bitcoin. It's called tax department, and of course it can impose tight(er) regulations.
• Otherwise, a simple 51% attack will do some work to destroy the reputation.

It will depend on whether the state wants to destroy Bitcoin or to acknowledge it and welcome it as part of the system.

[d5000]

• Otherwise, a simple 51% attack will do some work to destroy the reputation.

This remains to be seen. I mentioned this scenario briefly in my last post: if there's not much damage (in terms of double spends) done, then it could be possible that BTC recovers relatively fast, as did most of the altcoins which were 51% attacked until now. And if this happens, then all the effort of the state was in vain, and even such an attack would be expensive.

A big dictatorship like China or Russia could perhaps be tempted to perform this type of attack anyway because they could try to hide the attack, as they don't need to care that much about transparency. But for any government of a democratic state such an attack would be basically suicide - why should people pay with their taxes for an attack on a tool that e.g. 10% of the voters already use? This would be a major scandal. In China or Russia it could work to hide it, but there are risks that such an attack could destabilize the government if there are traitors leaking the attack. I believe even in the most aggressive and authoritarian of the big dictatorships, Russia (and much less China) they aren't that dumb to waste billions on that.

The only actual constellation I can imagine is an intelligence department of a dictatorship performing the attack, but sponsored in secret by several other countries (every government contributing with, let's say, less than $1 billion, which perhaps can be hidden). But even there, all participants would be at risk to destabilize.

Thus a FUD & Regulation attack is much more promising. But if they go too far the attack could impact in public opinion against the governments and could lead to resistance, like we've seen in the European Parliament. The more Bitcoiners we are, the less likely is such an attack. I think for a completely destructive attack of this kind thus it's already late.

We could also speculate: was Ordinals such an attack?

The idea itself is not bad for a malicious entity - create some service on top of Bitcoin that lowers the utility of the chain for most users because it clogs the blocks, but leads to profits for a sub-group like the NFT investors and can even return the cost of the attacks from "greater fools" (like those buying ORDI at $60 or more). However, due to the way such "trends" work, sustaining such an attack would be very expensive too. And we can already say that it probably failed, because Bitcoin became even more popular. The long term solution to make such a "spam attack" much more unlikely would be - second layers (sidechains, rollups and LN).

[Rabbitqt]

I hear a lot about how safe bitcoin is. And 51% attacking it for a private person/ company does not make any sense and is unlikely.

But what would happen if a state wants to attack Bitcoin and uses a vast amount of compute budget (in the billions). What would happen?

Bitcoin could potentially face challenges from state-level agents capable of deploying significant computational power. While current computing resources make it possible to perform a 51% attack on the network, the scale and sophistication of such attacks could increase if state actors were involved. However, Bitcoin's decentralized nature and the ongoing efforts to enhance its security protocols provide some level of defense against such threats.

[Pmalek]

There is no easy way to attack Bitcoin and take control of it - become the powerhouse that makes the most money from it. That investment wouldn't pay off in the end.

It's a bit different if your goal is try to destroy Bitcoin or inflict as much damage to it as possible. In that case, the most vulnerable category would be miners. You would need a global action on a massive scale against mining companies. Stop their operations, confiscate their hardware, and stop the production of new blocks and confirmation of transactions.

Bitcoin can recover from that with a very bloody nose. The mining difficulty would drop, making other machines suitable to take over the process from those previously confiscated. But than the global police would have more powerful hardware than the honest miners.

All this is sci-fi and will never happen. Our leaders can't even make up their mind if it's ok to kill 35.000 innocent people or not, let alone come to an agreement and organize something on a massive scale like an attempt to destroy Bitcoin.

...not mentioning that since those billions would be taxpayers' money, if the government is a democracy, it will be the last thing (use of money) they'll approve before getting changed and the preorders canceled.

That taxpayer money could be used much more wisely. For example, on fighter jets and massive bombs intended to be dropped on a small piece of land. You know, for democracy and preserving freedom.

[Grace333]

The reputational damage for Crypto currencies  will have the biggest impact, if this succeeds. ( People will lose trust in Crypto currencies in general)

Exactly what was going on in my mind. If this attack is executed, it gives a negative feedback to not only users of crypto but to everyone across the world. The reason many person dive into Bitcoin investment and not all altcoins is because of the security and reputation of Bitcoin, if that is broken then, no altcoin would be trusted anymore and that might mark the beginning of the downfall of cryptocurrency. However there are approximately 18,000 nodes running now, and if any attack can be executed it means the attacker must have about 9180 nodes which is quite a large number of nodes to set up.