All The Ways Bitcoin Can Fail

Moises Cassab
Sep 19 · 6 min read

Bitcoin has been the strongest cryptocurrency of the past decade. It has overcome every major obstacle since its inception including the CVE-2010 Bug, Mt.Gox, the BCH Fork Wars, and the Rise of Altcoins. This doesn’t mean we have solved every conceivable attack and that Bitcoin is a shoo-in with 100% odds of becoming the global SOV that will endure for millennia.

I’ll add the disclaimer that yes, there are solutions to most of the points I’ll mention; hopefully, none will occur, and, invariably, not everyone will agree with what you’re about to read.

This is my take on the biggest problems Bitcoin faces:


1- Mining power is centralized: Every part of the mining infrastructure is centralized.

A) Bitmain produces the majority of miners and backdoors that could grind block production to a halt have been found in them.

B) 8 mining pools control 85% of the Hash Power.

C) You could probably fit most (70%+) of mining power in one room. In fact, we already did it once, in 2016 a miner led HF was imminent and core devs alongside miners got into a room in HK and decided the fate of bitcoin.

D) Most of the hash power is in a country that is hostile to Bitcoin and freedom in general.

In short, the whole mining stack is extremely centralized.

The attacks that can be done with a majority has power are a chain reorg, 51% attack, Death spiral, eclipse attack and doing a POW algo reset opens up another can of worms.


2- Shorting and alternative chains distort game-theoretical considerations laid out by Satoshi. A key part of the game-theoretical consideration was the miners had a long term incentive to be good actors since their miners were tied to the success of bitcoin. This is no longer the case. Miners could attack the bitcoin chain, and short bitcoin while they do so, make an immediate profit and then mine another chain.


3- Untested security model, the fee market: Every 4 years the security budget of bitcoin halves in BTC terms. The main arguments against the chaos this would bring are the following:

A) We don’t need to worry about till 2140 when it hits 0. However, in practical terms, in 9 short years, the security budget will be down by 88% from the current 12.5BTC to 1.5BTC and in 17 years to 0.39BTC per block. A 97% reduction from where we are now.

B) The price will increase by more than 2x by the next halving. The problem with this thinking is that the bounty for meddling also increases with price. Additionally, following this logic would mean that in 9 halvings Bitcoin is worth more than the entire global GDP.

I can only see three ways to get out of this pickle.

A) Proof of Stake. Not gonna happen.

B) Violate the 21m limit and stop the halvings and keep mining rewards as they are now. Not gonna happen.

C) Take some of Satoshi’s coins and use them to fund inflation. Could see this happening.

D) Let a fee market develop. This is the most likely path, the problem with this is that if we want to have 12.5 BTC per block in fees we are looking at .0034BTC per Tx. or 340 bucks if bitcoin is worth 100K. Yet this isn’t the scary part, it’s the unknown behaviors that will develop.

What we do know is that we can’t have it both ways, miners MUST be paid, either in fees or subsidies. So, we will invariably end either with high fees or have other coins eclipse BTC in Hash Power… and that opens up the attacks in point #1.

Also, it would suck to have to pay fees and would render the LN too expensive for the majority of the world who will not pay a few days wages to lock and unlock their bitcoins.

Possible attacks: Other higher inflation coins have a higher hash rate that can do attacks like those mentioned in point 1 and high fees make the network unusable for the masses and remain a niche product.


3- Bitcoin dies a slow death due to lack of innovation: This one is both a pro and a con. Bitcoin will not change, and that is ok.

But we cannot deny that innovation has dried up. Since I bought bitcoin in 2013 there hasn’t been a substantial new application or use case from the users perspective. Yes, the nodes sync faster; yes, signatures are smaller… but the big innovations we were promised never materialized and some features even went bust.

  • ChangeTip shut down and was acquired by Airbnb
  • Colored coins were awesome but those went the way of the dodo and gave us the rise of ETH and alt-season via ERC-20. Also dead.
  • Counterparty promised smart contracts, but core unilaterally decided to change OP_RETURN from 80 bytes to 40 bytes.
  • Blockstream promised sidechains where bitcoin was the base layer and all the experimentation would be handled in the sidechains, it would be trustless and awesome. Now they’ve basically given up and are doing a permission federated sidechain where all the nodes are pre-approved by Blockstream and run on proprietary hardware.
  • Rootstock, a federated sidechain where you trust a few handpicked validators.

All that is left standing is Lightning and it will be interesting to see how it plays alongside a fee market. Also, Joseph Poon, one of the authors of the white paper, has since abandoned Bitcoin and works on Ethereum.

The risk here is that bitcoin never goes mainstream and dies a slow death.


4- Lack of privacy: Bitcoin at best is pseudonymous. If we were to face a crackdown by a large nation, all the satellite nodes and decentralization won’t matter if authorities can easily identify holders a beat them up with a wrench.


5- Centralized development: In the past year 70% of commits were done by 10 individuals and 97% of nodes are core. We can pretend everybody checks all the 300K+ lines of code and there was consensus on all the changes. But realistically, we follow what a few rockstar core devs recommend.

The odds of some insidious code or bad decisions by the core team of developers is non-zero. See CVE-2010–5139 + CVE-2013–3220, and these were just honest mistakes, not a targeted attack.


6- Quantum computers: As it stands “Bitcoin is not quantum secure, period” and quantum computing by no means is a pipe dream. Quantum computers exist, work, and are extremely underpowered. However, this doesn’t mean they won’t get better in the coming years.


7- It can’t scale to mainstream: At its peak bitcoin has processed 2763 Tx in a single block. This would imply bitcoin can process 145,223,280 per year, so about 0.02 transaction per year per person. Thus, even if we get the cheap fees and Lightning works flawlessly people can only open and close one channel. It would take 96 years to onboard the world.

Even if bitcoin were just used as a SOV, that is, if people just bought BTC and did a single Tx and never moved again it would take 48 years to onboard the world.

It remains niche, nerd money.


8- Black swans, unknown unknowns: Anything not on this list like a big asteroid hitting the Earth or a murderous AI. The things we just don’t know.

So as promised, everything that can kill bitcoin is here.


Credit to my good friends, Allan, Mario Romeo, Matt, Josu,y extra thanks to Santiago who made this legible and Jordan McKinney for his awesome post on BTC security. Also, Nassim is a legend and you should immediately read Antifragile.

Feel free to add or subtract ideas to the list in the comments section or on twitter @cryptonaut55

Moises Cassab

Written by

Techie, Bitcoiner, Simulationist, Manufacturing. Forward Looking Statements and Black Swans

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade