What this covers

Neha Narula, director of the Digital Currency Initiative at MIT, sits down with Sean Carroll to examine money itself and the technological systems reshaping it. The conversation begins with a foundational claim: money is not valuable because of any physical property of its medium, but because of a collective story people tell and enact. From there, Narula traces how Bitcoin, introduced in 2008 by Satoshi Nakamoto, became the first system enabling peer-to-peer digital payments without banks or governments by combining private key cryptography with decentralized mining. The discussion moves outward to explore what blockchain technology promises—permissionless financial access and a more decentralized web—alongside what it actually delivers in practice.

The conversation covers substantial technical and economic ground. Narula explains how blockchain works as an append-only ledger, how mining functions as both validation and incentive, and why scaling poses real constraints on transaction fees. She identifies genuine use cases: cross-border remittances and wealth preservation in countries facing capital controls or hyperinflation. But the bulk of the exchange dwells on contradictions and limits. Decentralization is hard to use because no company profits from polishing the interface. Cryptocurrency exchanges like Coinbase undermine self-custody. A lost private key means permanent loss of funds. Censorship-resistant platforms become havens for toxic content. Blockchain is the wrong tool for voting. Market caps are easy to fake. Narula neither dismisses the technology nor oversells it; instead, she traces where the vision meets friction and where real risks remain unsolved.

Sharpest takeaway

Narula argues that money is fundamentally a shared story rather than something with intrinsic value, and that blockchain-based cryptocurrencies—by enabling peer-to-peer digital payments without banks—could expand financial access and decentralize the web, though the technology remains immature and carries real risks.

  • Even 'intrinsically valuable' assets like gold derive worth from collective convention, not inherent properties
  • Bitcoin was the first system to enable digital payments without a bank via decentralized mining and public key cryptography
  • Decentralization promises broader access and individual control but faces usability, security, and abuse problems

The argument · threads9 threads · 35 claims
0.79

Cryptocurrency enables permissionless access to financial services for the billions currently unbanked or excluded from traditional systems.

4 pointscentrality 5/5
  • The truly groundbreaking aspect of cryptocurrency is not slightly lower fees but permissionless access: anyone with an internet connection can start using it without a bank signing them up, without KYC or AML checks, without a minimum deposit or specific identification—similar to how the internet itself is permissionless.

    “if they can get an internet connection, they have access... they don't need to be KYC'd or AML'd, they don't have to have a minimum deposit size or a certain type of identification, they can just start doing it... what we call permission less, which is the Internet.”

  • Bitcoin, introduced in 2008 by the pseudonymous Satoshi Nakamoto, was designed for peer-to-peer digital payments without any bank, government, or other institution facilitating those payments—the first time people could make digital payments without a bank.

    “the idea behind it was peer-to-peer payments, without any bank or government or other institution, facilitating those payments. Peer-to-peer digital payments.”

  • Roughly two billion people in the world are 'unbanked,' lacking access to digital financial services like loans, savings accounts, and digital payments, because reaching them is expensive and assessing their creditworthiness is difficult, leaving them unable to use services that could improve their quality of life.

    “the general statistic, that's thrown around is that there are about two billion unbanked people in the world.”

  • The world is tremendously unequal, with roughly 12 very large investment banks effectively running the financial system, moving trillions of dollars in value daily, while ordinary individuals face decisions over thousands of dollars that are mere rounding errors to those institutions.

    “about 12 really large investment banks basically, kind of run things. They move around trillions and trillions of dollars worth of value every day.”

0.71

Money and value derive from collective agreement and shared stories rather than intrinsic properties of any object or medium.

5 pointscentrality 4/5
  • Economists define money by three properties: it is a unit of account (used to price things), a medium of exchange (used to transact and have an economy), and a store of value (used to save and preserve wealth over time).

    “economists define money, as having three qualities. It is a unit of account, we use it to price things... as a medium of exchange... And then the third... property that defines money is as a store of value.”

  • Gold does not have intrinsic value equating to its market worth; while it has some industrial and decorative uses, its role as a measure of wealth is itself another story society tells, no more inherently valuable than fiat currency—'it's turtles all the way down.'

    “Why is gold worth something? It's turtles all the way down is what I got to say. Gold does not actually... Doesn't have intrinsic value which equates to the amount that it's worth in our society today.”

  • Money has value not because of any intrinsic property of the object, but because we collectively tell each other a story that it is worth something and interact as though that story is true—we agree to use objects, paper, or ones and zeros in databases as accounts of value and use them to transact.

    “money is a story. Money is a story that we tell ourselves... It's worth something because we tell a story to each other about why it's worth something, and we all sort of believe in that story.”

  • There is no cut-and-dried thing called intrinsic value; different people value the same things—including bread, water, or art—very differently, and society negotiates these differences through prices, so even seemingly essential goods lack a fixed inherent worth.

    “Things have different values to different people, things like art... this is a very incredibly different value to different people. We try to negotiate that using prices and value.”

  • Market capitalization is computed by taking the price of the last share sold and multiplying it by the total number of outstanding shares, which can produce absurdly inflated valuations—as demonstrated by issuing a billion 'Neha Shares' and selling a few for a dollar each to claim a billion-dollar market cap.

    “it pokes fun at crazy high market caps of startups or cryptocurrencies... It's the last share you sold, and you multiply that times the number of outstanding shares.”

0.71

Decentralized systems struggle with usability because no company profits from investing in user experience and interface design.

5 pointscentrality 4/5
  • Decentralized systems are very hard to use because, without a giant company in the middle making money, there is no one to pay for the UI designers, experienced engineers, and people needed to make the product polished and usable; crowdsourcing rarely fixes this because user-experience work is not what volunteer coders find interesting, as illustrated by Linux being less user-friendly than Mac OS.

    “this is a big problem with decentralized systems. They are very hard to use, and if there's no giant company sitting in the middle of it, making a ton of money, who's paying for the UI designers and these are experienced engineers...”

  • When you hold cryptocurrency through a company like Coinbase, you are not actually holding it yourself—the company holds it and you have an account with them, which is more like a bank account than the self-custody peer-to-peer ideal that crypto originally promised.

    “when you hold cryptocurrency through a company like CoinBase, you're not actually holding it yourself, they're holding it. You have an account at CoinBase, it's a little bit more like a bank account.”

  • Cryptocurrency technology is still in its infancy and not ready for prime time—comparable to clunky, slow, gas-guzzling early-1900s cars that horses could outpace—but it is improving rapidly and solutions to its current problems are in sight.

    “in the early 1900s cars were very good clunky, very slow. Used a lot of gas. Horses were faster than cars, but they got better very, very, very quickly. And I think cryptocurrencies are kind of the same way.”

  • In decentralized cryptocurrencies there is no company to reset a forgotten password, so losing your private key—the secret number controlling your funds—means the money is permanently lost and unrecoverable, and if someone steals it they can steal your money; an estimated $20 billion in crypto has been lost this way.

    “if you lose your private key, nobody's gonna be able to give that back to you, and nobody can reset your account for you. That's kind of the point. And so, yeah, you're kind of out of luck.”

  • Public key cryptography, invented in the 1970s (with Ron Rivest contributing at MIT), gives each user a private key never shared with anyone and a public counterpart; the private key signs or decrypts while the public key validates signatures or encrypts for the owner—fundamentally different from passwords, which must be transmitted to a server.

    “this is called public key cryptography, and it was actually invented in the 70's and it was invented at MIT. A professor I work with occasionally named Ron Rivest was pretty critical in its invention.”

0.65

Blockchain is a tamper-proof ledger structure where transactions form an append-only chain that everyone can verify and no single actor controls.

4 pointscentrality 4/5
  • Bitcoin mining involves gathering transactions into a block and repeatedly trying random numbers until one produces a hash with the required number of leading zeros—finding the pre-image of a hash—which is computationally hard to do but easy for others to verify; successfully solving it validates the block and lets the miner create new Bitcoin according to a fixed schedule.

    “it's actually something called 'Finding the pre-image of a hash'... your computer's trying different strings over and over and over again, to see if they produce the right kind of fingerprint... a short string with a lot of zeros at the front.”

  • A blockchain is an append-only ledger data structure copied across every node in a network; transactions are grouped into blocks and each block points to the one before it, forming a chain, which makes it tamper-proof because altering past transactions would break the chain of pointers in a way everyone would immediately notice.

    “each block points to the one before it. So it's like this list... a link list in Computer Science terminology. And what this serves to do is it makes the blockchain tamper-proof, you can't go back in time and fiddle with the transactions, because if you do, you're gonna mess up all the pointers.”

  • Through 'forking,' someone can divide a cryptocurrency network and create a copy with a new ledger and new blockchain, analogous to biological speciation; this has happened to Bitcoin several times, producing variants like Bitcoin Cash and Bitcoin Gold.

    “there's something called forking, basically someone can decide to divide the network up and basically create a copy of Bitcoin, new network, new ledger, another blockchain.”

  • A hash function maps a large amount of data onto a short, unique-looking fingerprint; even though many inputs can map to the same output and the function is publicly known, it is extraordinarily difficult to work backwards from a hash to an input that produces it—this one-way property is what makes mining hard.

    “a property of a hash function is that it is extraordinarily difficult to find something... given the short string, the hash the fingerprint, it is very difficult for me to work backwards from that, to something that will hash to that value.”

0.52

Real cryptocurrency use cases concentrate in cross-border remittances and wealth preservation for people facing capital controls or hyperinflation.

4 pointscentrality 3/5
  • The most moving real-world cryptocurrency use cases fall into two categories: cross-border remittances, where moving money between countries via Bitcoin is much cheaper than conventional means, and wealth preservation, where people in countries with capital controls or hyperinflation use crypto to get their accumulated wealth out before it disappears.

    “One category is I earn money in one country and I want to put it in another country, and by using something like bitcoin I'm able to do that much more cheaply... The other use case... people from countries where their currency is under capital controls, or it's getting hyper-inflated, and this offers them a means of taking the wealth... and getting it out before it disappears.”

  • In its early years Bitcoin was a toy that people on the internet were playing with, and its value derived from a story they told themselves—that it was an unhackable form of money based purely on mathematics that no one could take from you.

    “This was an unhackable form of money, this was money that no one could take from you. It was purely based on mathematics, it was a very pure form of internet money.”

  • Storing value is genuinely difficult because value is not static—stock portfolios can crash, currency exchange ratios can shift dramatically, and hyperinflation can wipe out wealth, so even cash under a mattress is not guaranteed to be worth the same tomorrow.

    “the world is a bit of a harsh place and it is actually very difficult to to store value, and to really ensure that it's gonna be worth the same amount tomorrow.”

  • The first exchange of Bitcoin for something of real-world value occurred when someone bought two pizzas for 40,000 Bitcoin, establishing an early implied price (pizza cost divided by 40,000), an amount now worth tens of millions of dollars.

    “If you send me two pizzas I will send you 40000 Bitcoin and this is what's known as the first exchange of Bitcoin for something of actual value.”

0.52

Cryptocurrency and decentralized platforms create tension between censorship resistance and the ability to remove harmful or toxic content.

4 pointscentrality 3/5
  • Digital payments and the ability to issue tokens or shares in oneself are connected because both are about disentangling the instruments of finance from trusted institutions and the intense regulation around them—breaking that apart and reimagining how financial activities might work if they weren't so encumbered.

    “they're all about sort of disentangling the instruments of finance from trusted institutions, and the very intense regulation that goes around those institutions. It's about breaking that apart and reimagining how we might do these things if we weren't so encumbered?”

  • Censorship-resistant federated platforms like Mastodon, where you can simply join another instance if kicked off one, are harder to censor and thus become breeding grounds for the toxic content that gets removed from centralized platforms like Twitter—revealing the trade-off that decentralization protects speech we may collectively want censored.

    “these things are breeding grounds for the type of things that get kicked off of Twitter, which are not really very savory sorts of things... it's sort of the counterpoint to censorship resistant technology that means that you really can't censor things that may be even as a society we would want censored.”

  • Centralized internet platforms like Facebook have monopoly-like power over public discourse with algorithmic bias and censorship, and users feel trapped because even if they dislike a platform, that's where all their friends are, so leaving means losing access to their social network.

    “people are very worried about the power that these centralized platforms have over our society... there's algorithmic bias in the ways that they operate, their censorship, and people sort of feel trapped.”

  • Applying crypto's user-ownership ideas to the web suggests data portability—being able to move all your content out of a platform like Facebook into a competitor at the click of a button—and interoperability, so that even on a different social network you could still interact with people who stayed on the old one, recreating the internet's original interoperable-protocol vision.

    “I should be able to get it out whenever I want, at the click of a button, and I should be able to move of it... even if I'm on a different social network, I should still be able to interact with the people who stayed behind on the other social network.”

0.46

Bitcoin's consensus mechanism uses mining to secure the network while making attacks economically irrational despite its open-entry design.

3 pointscentrality 3/5
  • Mining moved from individual laptops to specialized ASIC (application-specific integrated circuit) machines that solve the hashing problem millions of times faster, so individuals can no longer effectively mine Bitcoin without such hardware; this also led cryptocurrency miners to buy up GPUs, competing with gamers and driving prices up.

    “now there are these machines called ASICs application specific integrated circuits that can solve this problem millions of times faster than your laptop can and basically, you can't mine in the Bitcoin network effectively unless you have one of these machines.”

  • Bitcoin's high fees stem partly from the fact that every node stores a copy of the ledger, so a single transaction gets stored thousands of times around the world, incurring real costs in bandwidth, disk space, and CPU to validate—making transactions reasonably pricier, though coming solutions could make this much cheaper.

    “everyone's storing a copy of this ledger. So when I create my one little transaction, it gets stored thousands of times around the world... It costs money to pay for the bandwith to move that around, to pay for the disk space to store it, to pay for the CPU to validate it.”

  • Unlike conventional security models that build a perimeter firewall to keep hackers out, Bitcoin has no perimeter—anyone can join at any time, effectively inviting hackers in—but its use of digital signatures and the mining game means hackers cannot spend others' money and are instead incentivized to participate productively rather than attack the system.

    “With Bitcoin, there's no perimeter, anybody can get in at any time. Bitcoin invites the hackers in, so to speak... but it's designed in such a way that by using digital signatures and by creating this strange game of mining, the hackers won't be able to spend my money and they will be incentivized to participate in a productive way instead of attacking the system.”

0.44

Cryptocurrency technology is nascent and improving rapidly, likely to enable unanticipated applications as a general-purpose platform for value transfer.

5 pointscentrality 2/5
  • The goal is to create an 'Internet of Value'—doing for the transfer of value what the internet did for information—by building platforms and infrastructure that let people easily build applications to move, transform, and contract value, including micropayments, without anticipating exactly what uses will emerge.

    “we wanna create what we call the Internet of Value so... What the internet did for information, we wanna do for the transfer of value.”

  • Just as early internet enthusiasts couldn't explain or predict what the web would be good for, the most important future applications of crypto and blockchain are likely to be unanticipated; the technology is best understood as a platform on which many things will be built, some surprisingly valuable.

    “there's something that we can do here that we couldn't do before, and I don't know how we're gonna do it, but it's a platform. People will build a million things on top of it, and some of those things will be super amazing and cool.”

  • Making it orders of magnitude easier to charge for things via micropayments has a dystopian downside: things currently free because charging isn't feasible—park benches, water fountains, public infrastructure—could become metered, and pricing could become unfairly variable (like dynamic toll or airline pricing), eroding the commons.

    “right now, there's a lot of stuff that we just get for free because it's not really feasible to charge for it... if we create a world toward taking payments and charging for things, it just becomes orders of magnitude easier than what that might do. That's a little scary.”

  • Whereas Bitcoin is a ledger of transactions, Ethereum functions more like a global computer that runs smart contracts—little programs thrown onto the blockchain that then execute—with one very popular type of program letting anyone create their own token.

    “Ethereum is kind of more like a global computer... it handles something called Smart contracts, which are like little programs, so you can write a little program and throw it on the Ethereum blockchain and then that little program we execute.”

  • Consumers think they don't pay credit card fees, but merchants pay a fee around 2.7% and pass it on through the prices of goods, so consumers effectively bear the cost; cryptocurrencies might push such fees somewhat lower, though that alone isn't groundbreaking.

    “we don't think we pay fees for credit cards, but actually we're paying for it in the goods that we buy. The fee, around 2.7%.”

0.20

Blockchain is the wrong solution for voting because endpoint security vulnerabilities vastly outweigh any benefits from distributed ledger structure.

1 pointcentrality 2/5
  • Blockchain technology is the wrong solution for electronic voting because the real vulnerabilities lie in the endpoints—the phone or computer, internet connection, and entire software stack—where any bug can compromise a vote, and blockchain does nothing about that endpoint security; paper is hard to hack at scale while electronic systems vastly enlarge the attack surface, so caution is warranted and unvetted startups should not build mission-critical voting infrastructure.

    “blockchain technology doesn't help with any of that endpoint stuff. It doesn't help make sure that my... I didn't install some weird piece of software that's sitting there fiddling with my vote... And so, I just don't think it's solving the right, the real problems.”