Prologue. The Plane Never Landed
illustrationIn the summer of 1974 the physicist Richard Feynman gave a commencement address at Caltech and told the graduates about some islands in the South Seas. During the war, military planes had landed there and unloaded canned food, cloth, tools and other treasures the islanders had never had. The war ended and the planes flew away. The islanders wanted them back, so they did everything the way it had been done before. They cleared a runway and lit fires along it, built a wooden hut and put a man inside with two pieces of wood on his head for headphones and bamboo sticks for antennas, and then they sat down to wait. By Feynman’s account, they did everything right: the form was perfect, and everything looked exactly as it had before. The planes didn’t land.
Feynman called this cargo cult science and warned the future scientists against it, since they too could end up with research that has every outward sign of the real thing, the papers, the graphs and the citations, but no cargo. About money he said not a word in that speech.
illustrationOn the fourteenth of July 2010, Habr, the big Russian tech site, ran a post titled “Bitcoin (฿): A Peer-to-Peer Cryptocurrency.” Its author, alizar, described version 0.3.0 of a program that a person calling himself Satoshi Nakamoto had put online a year and a half earlier. The post promised that the new currency “cannot be devalued by the policy of any country’s Central Bank” and that the issuing of coins would stop once there were 21 million of them, and along the way it honestly admitted that bitcoins were accepted as payment “by roughly ten merchants.” It collected 175 comments.
I read the post and installed the wallet on my MacBook Pro, the first fifteen-inch one Apple released on an Intel processor. In those days you mined coins right in the wallet. There was a menu item that switched on generation, and the processor started running through numbers in search of the one that would fit the next block of records. For every block it found, the network paid out 50 bitcoins. I switched generation on and forgot about it. The laptop got as hot as an iron, and after a while it found two blocks, so the wallet ended up holding 100 bitcoins. All together they were worth less than dinner at a café, and I forgot about them too.
Fifteen years later, analysts at Chainalysis calculated that in 2025 alone at least $154 billion had flowed into wallets linked to criminal activity, while in the same year the FBI received more than 180,000 complaints from Americans about crypto fraud, totalling more than $11 billion. On the platform pump.fun anyone can set up their own currency in a couple of minutes, and when the firm Solidus Labs examined seven million coins launched there over a little more than a year, more than 98 percent of them turned out to be traps in which the creator pumps up the price, sells his share to those who believed, and vanishes. The ICO era, when tokens were sold on the promise of a future product, looked only slightly better, because back in 2018 analysts at Satis Group classified about three quarters of large projects as scams.
The industry’s answer is that criminal money makes up about one percent of all crypto turnover, and that figure is honest. Only the turnover consists almost entirely of speculation, where the same coins change hands many times a day, and the fraud on exchanges, ICOs and memecoins runs through perfectly clean wallets and never enters that percentage at all.
I took part in this story myself. ChronoBank, the project where I first wrote the architecture and then served as chief technology officer, raised about $5.4 million in its ICO in the winter of 2016–2017. I crossed paths directly with the organizers of other ICOs, Russian and foreign, and with the creators of genuine pyramid schemes, most often in Thailand, which became for such people what Las Vegas became for gamblers.
archiveSatoshi’s invention itself was real. For more than twenty years cryptographers had tried to work out how strangers who don’t trust one another could keep one shared record of who has how much money, with no bank in the middle, and every time they ran into the fact that someone would spend the same coin twice. Satoshi made the participants’ computers compete at a seemingly pointless search through numbers, so that rewriting the payment history became more expensive than taking part in it honestly, and in 2010 I looked at this construction with the delight an engineer feels at someone else’s solution that he could have come up with himself but didn’t.
The properties of this record were appreciated fastest by people far removed from cryptography. A bitcoin payment, once sent, cannot be reversed, the bank doesn’t see it, a border means nothing to it, and to issue a new coin you don’t have to ask anyone. To an honest person with a bank card all this looked like a curious toy for geeks, but someone the bank had turned away, someone the police were looking for, or someone who wanted to sell something forbidden got a way out he had never had before.
Money in this record belongs to a key, a long secret number, and a person owns it only as long as he alone knows that number. Whoever loses the key leaves the money in the record forever, owned by no one, and whoever steals the key becomes the full owner, because the record knows no other owner. Many years later everything I had seen in crypto fit, for me, into three words, “nobody owns money,” and they have a double meaning. In the first sense, the owner of crypto is whoever knows the key at that moment, and most of the time he cannot be found. In the second sense, there is nothing behind most coins except numbers on a screen, and those numbers are worth whatever the next buyer will pay for them.
In February 2011, seven months after the post on Habr, a shop called Silk Road opened in the hidden part of the internet that you reach through a special browser, selling drugs for bitcoins. Satoshi’s invention had found its first customers who truly needed it.