Mike Fluff Press

Answers

The questions people most often put to search engines and chatbots about Bitcoin, crypto and programming with AI. Each answer is drawn from one of the books and leads to the chapter that tells the story in full.

Nobody Owns Money

11 questions

  1. 01Who is Satoshi Nakamoto?Satoshi Nakamoto is the name signed to the nine-page bitcoin paper sent out on 31 October 2008, and the author of the program launched on 3 January 2009. Who is behind it is unknown: he left the forum in December 2010, stopped writing in spring 2011, and the roughly one million bitcoins attributed to him have never moved. None of the theories about his identity has been proven, and in 2024 the High Court in England rejected Craig Wright’s claim.
  2. 02What is the double-spending problem?The double-spending problem is that a digital coin can be copied, so its owner can pay two sellers with the same file. For years the only fix was a middleman who keeps a shared list and crosses out spent coins, which is just a bank inside a computer. Bitcoin replaced the middleman with a network where rewriting the payment history costs more than taking part in it honestly.
  3. 03How does proof of work work in bitcoin?Proof of work is a certificate that electricity has honestly been spent on something: a computational puzzle is hard to solve but instant to check. In bitcoin such a puzzle stands at the entrance to the shared ledger of payments; across the whole network a solution is found about once every ten minutes, and whoever finds it first gets new coins. History can be rewritten only by redoing the work faster than all honest participants combined.
  4. 04What is the Bitcoin Pizza?The Bitcoin Pizza is the two Papa John’s pizzas that programmer Laszlo Hanyecz paid 10,000 bitcoins for in May 2010. It is counted as the first real-world purchase made with bitcoin, although the pizzeria was paid in ordinary dollars: a student nicknamed jercos paid about 25 dollars for the order and kept the coins. Every year on 22 May the crypto community marks Bitcoin Pizza Day.
  5. 05Why were drug dealers and ransomware gangs bitcoin's first real customers?Bitcoin's properties solved their main problem. A payment cannot be reversed, the bank does not see it, it knows no borders, a wallet takes no passport, and a million dollars fits into twelve words. An honest person with a bank card in 2011 had almost no use for any of this, but a trade that lived on cash needed it badly, and bitcoin found its first real demand on Silk Road.
  6. 06Is cryptocurrency mostly used for crime?The industry's answer is that criminal money makes up about one percent of all crypto turnover, and that figure is honest. But the turnover is almost entirely speculation, where the same coins change hands many times a day, and the fraud on exchanges, ICOs and memecoins runs through clean wallets and never enters that percentage. Crypto does have honest uses too, and nearly all of them involve digital dollars.
  7. 07How does a crypto Ponzi scheme work?A crypto Ponzi scheme pays old participants with new participants’ money and survives as long as more people come in than leave. Crypto gave these schemes money that a bank cannot freeze, and the word “blockchain”, behind which an ordinary database is easy to hide. The best-known cases in the book Nobody Owns Money are Sergei Mavrodi’s MMM, OneCoin and BitConnect.
  8. 08What is a pig butchering scam?Pig butchering, shazhupan in Chinese, is a scam in which the victim is courted in chat for weeks and then talked into putting money on a crypto platform where the balance grows on screen but can never be withdrawn. Payment is almost always in crypto, most often USDT, because such a transfer cannot be reversed. At the other end of the chat there is often a person held by force in a scam compound in Myanmar or Cambodia.
  9. 09Why did Terra and LUNA collapse?The stablecoin UST was held at one dollar by a second coin on the same network, LUNA: any UST could be burned for a dollar’s worth of LUNA. In May 2022 UST holders all rushed for the exit at once, the algorithm printed 6.5 trillion LUNA, and its price fell from 80 dollars to fractions of a cent. Within days Terra and LUNA, worth about 40 billion dollars on paper a month earlier, were worth nothing.
  10. 10What happened to FTX?In November 2022 Sam Bankman-Fried’s exchange FTX collapsed within a week: after a CoinDesk article about Alameda Research’s balance sheet, customers withdrew about 6 billion dollars in three days, and on 11 November FTX filed for bankruptcy. It turned out that customer money had sat for years in the accounts of Alameda, which the exchange’s code allowed to go negative. Bankman-Fried was found guilty on all seven counts and sentenced to 25 years.
  11. 11What are stablecoins and what are they used for?A stablecoin is a digital dollar on a blockchain: a token such as USDT or USDC that is meant to be worth one dollar at all times, because the company that issues it holds a reserve behind it. They are used most where the local currency is failing: in Argentina, Turkey, Nigeria and Venezuela people keep savings in them, and migrants send money home for cents. The other side is that a private company issues them, and it can freeze any wallet.
  1. 01What is vibe coding?Vibe coding is a way of making software in which you tell a model what you want, look at the result and paste the error messages back to it, without reading the code itself. Andrej Karpathy coined the term on 2 February 2025, and Collins named it word of the year for 2025. It works while the program is small and breaks once other people’s money and data sit behind the screen.
  2. 02Why is AI-written code cheap to write but expensive to maintain?Language models brought the cost of writing code close to zero: in autumn 2026 a million lines from the cheapest models costs about a dollar. The cost of keeping it stayed where it was, because every line remains an obligation that has to be read, checked whenever something next to it changes and carried along with the system. Writing became almost free, and not writing what is not needed became an expensive skill.
  3. 03Why do AI agents cheat on tests?For an agent, the test is the reward: models were trained for years to write code that passes checks, and when fitting the result to the test is easier than solving the task, a capable enough system will sooner or later find that path. There is no malice in it. It is Goodhart’s law at work: when a measure becomes a target, it ceases to be a good measure.
  4. 04Where will senior developers come from if AI writes the code?Nobody has a ready answer yet. The small tasks on which beginners used to build their skills for years are now done by agents, and the ladder to mastery has lost its bottom rungs. My bet is that a beginner should sometimes fly by hand without the autopilot, ask the model “why” instead of “do it for me”, and learn by taking apart the code agents write.