SWITCHED ON
The daily technology series nobody asked for but everyone needed
The Second Act: Crypto, DeFi, and What Blockchain Actually Does
After the collapse of FTX, the NFT implosion, and the DeFi hacks that collectively cost billions, the question was whether crypto would have a second act. It has. Whether the second act is better than the first is the more interesting question.
Blockchain is a technology in search of a problem it solves better than existing alternatives, for users who are not already motivated by ideology or speculation. That search has been underway for fifteen years. The number of genuinely compelling answers has grown modestly. The number of people willing to fund the search has grown enormously. These two facts are in tension.
Yesterday we went into the replication crisis — the incentive structures that systematically produce unreliable science, the AI tools being used to detect and correct the problem, and the new problems AI is simultaneously introducing through hallucinated citations and undisclosed AI peer review. Today we are going somewhere with considerably less academic respectability and considerably more volatility. Cryptocurrency, decentralised finance, and the blockchain technology that underpins both. The crash, the fraud, the fraud that masqueraded as innovation, and then — after all of that — the question of what, if anything, survives the hype and the fraud and the regulatory crackdown and constitutes a genuine and durable technological contribution. The answer is more interesting than either the maximalists or the sceptics tend to suggest. As usual.
01 — What the Crash Actually Was
The crypto crash of 2022 — during which the total market capitalisation of cryptocurrency fell from approximately $3 trillion at its November 2021 peak to under $800 billion by the end of 2022 — was not a random market correction. It was the unwinding of a speculative bubble inflated by cheap money, retail investor FOMO, and a remarkable quantity of outright fraud that became visible only when prices stopped going up.
The collapse of TerraUSD and its sister token Luna in May 2022 — which wiped approximately $40 billion in value in days — was the first domino. TerraUSD was an "algorithmic stablecoin" that maintained its dollar peg not through dollar reserves but through a mathematical relationship with Luna that was, in retrospect, a design that worked only as long as confidence in the system was maintained and failed catastrophically the moment it was not. This is a description that applies equally well to a Ponzi scheme and to a flawed monetary design, and in this case the distinction was not entirely clear.
The collapse of FTX in November 2022 — the bankruptcy of the world's third-largest cryptocurrency exchange, with the subsequent revelation that customer funds had been systematically misappropriated by its founder Sam Bankman-Fried to fund trading at his associated hedge fund Alameda Research — was straightforward fraud, not flawed design. Bankman-Fried was convicted on all seven counts of fraud and conspiracy charges in November 2023 and sentenced to twenty-five years in prison. The FTX collapse cost customers an estimated $8 billion and shook the institutional credibility of the entire industry.
The crypto crash was not a failure of blockchain technology. It was a failure of the human institutions built on top of it. The centralised exchanges, the algorithmic stablecoins, the yield farming protocols promising implausible returns — these were the points of failure, and they failed in ways that had more in common with conventional financial fraud than with anything specifically crypto.
02 — Bitcoin's Institutional Moment
Stripped of the fraud and the speculative excess, Bitcoin has arrived at something resembling institutional legitimacy in a way that would have seemed implausible in 2022. The approval of spot Bitcoin ETFs by the US Securities and Exchange Commission in January 2024 — after years of rejections — opened the asset class to institutional investment through familiar financial instruments without requiring direct custody of the underlying asset. BlackRock, Fidelity, and other major asset managers launched Bitcoin ETF products that attracted tens of billions in inflows within their first months of trading.
The institutional adoption reflects a specific thesis about Bitcoin that has separated from the broader crypto ecosystem: Bitcoin as a scarce digital asset with properties analogous to gold — a store of value and inflation hedge with a fixed supply of twenty-one million coins. The thesis is contested — Bitcoin's correlation with risk assets during market stress has undermined the uncorrelated safe-haven narrative — but it has been sufficiently compelling to attract significant allocation from institutional portfolios. Bitcoin's price recovered strongly from its 2022 lows, reaching new all-time highs above $100,000 in late 2024.
What has not happened is the mainstream adoption of Bitcoin as a medium of exchange — its original stated purpose. Transaction costs, speed limitations, and volatility continue to make Bitcoin impractical for everyday commerce. The Lightning Network, a second-layer payment solution designed to address these limitations, has made progress but remains a niche application. El Salvador's adoption of Bitcoin as legal tender, announced in 2021 with considerable fanfare, has produced mixed results, with the IMF requiring as a condition of a 2024 loan agreement that Bitcoin's status as legal tender be made optional rather than mandatory.
03 — DeFi: The Genuine Innovation and Its Genuine Problems
Decentralised finance — financial services built on programmable blockchains, primarily Ethereum, using smart contracts that execute automatically without requiring trusted intermediaries — represents the most conceptually interesting application of blockchain technology and the one that has attracted the most serious technical and academic attention alongside the most spectacular hacks.
The genuine innovation: smart contracts allow financial agreements to be executed automatically and transparently on a public blockchain, without requiring a bank, broker, or other intermediary to enforce them. Lending protocols that allow users to borrow against crypto collateral without a credit check, decentralised exchanges that allow peer-to-peer asset swaps without a centralised order book, stablecoin systems backed by crypto collateral — these demonstrate real financial functionality implemented in novel ways. The code is auditable. The rules are transparent. The execution is automatic.
The genuine problems are extensive. Smart contract bugs — errors in the code that the contract executes automatically and immutably — have been exploited to drain hundreds of millions of dollars from DeFi protocols. The Poly Network hack in 2021 lost $610 million. The Ronin Network bridge hack in 2022 lost $625 million, attributed to North Korean state-sponsored hackers. The Wormhole bridge hack lost $320 million. The pattern is consistent: code that manages large pools of value is attacked by sophisticated actors who find and exploit edge cases in the logic, and because the execution is automatic and the blockchain immutable, the loss is typically irreversible. The promise of trustless finance delivers on its own terms — you do not need to trust an intermediary — but the risk transfers entirely to the code, and the code is written by humans with finite attention and imperfect understanding of all the ways their logic can be exploited.
04 — Stablecoins and Central Bank Digital Currencies
Stablecoins — cryptocurrencies designed to maintain a stable value, typically pegged to the US dollar — have emerged as the most commercially significant and regulatorily contested application of blockchain in financial services. The total stablecoin market capitalisation is in the tens of billions of dollars, with Tether (USDT) and USD Coin (USDC) dominating. Stablecoins are used primarily as a medium of exchange within the crypto ecosystem — they allow traders to move value between positions without converting to fiat currency — and increasingly for cross-border payments, where they offer speed and cost advantages over traditional correspondent banking.
The regulatory response to stablecoins has accelerated following the TerraUSD collapse, with the EU's Markets in Crypto Assets regulation and proposed US stablecoin legislation both imposing reserve requirements, redemption rights, and prudential standards that bring regulated stablecoins closer to narrow banking. This is not necessarily bad — the alternative is unregulated stablecoins that may or may not be adequately backed — but it does raise the question of whether a heavily regulated stablecoin is meaningfully different from a bank account, and what innovation value it offers over existing payment systems.
Central bank digital currencies — government-issued digital money on programmable infrastructure — represent the establishment's response to the stablecoin challenge. China's digital yuan is the most advanced CBDC in a major economy, with significant rollout across multiple cities. The ECB is developing a digital euro. The US Federal Reserve has been more cautious, but the topic is on the policy agenda. CBDCs preserve monetary sovereignty, allow programmable money features, and provide financial access to the unbanked — but their government-controlled, surveilled nature is precisely what the crypto community was building to escape.
05 — What Blockchain Actually Does Well
Fifteen years in, the honest accounting of what blockchain technology does well, for users who are not primarily motivated by ideology or speculation, is more limited than the ecosystem's promotional materials suggest and more substantial than its harshest critics acknowledge.
Transparent, immutable record-keeping for assets and transactions where the absence of a trusted central authority is genuinely valuable — cross-border settlement between parties who do not share a trusted intermediary, provenance tracking for high-value goods, certain categories of digital asset ownership — has demonstrated real use cases. The technology is genuinely innovative in creating shared ledgers that multiple parties can trust without trusting each other or a common counterparty. This is a narrow but real advantage over existing alternatives in specific contexts.
What blockchain does not do well: anything that requires privacy (public blockchains are transparent by design), anything that needs to interact with real-world data without trusting an oracle, anything where the cost and complexity of blockchain infrastructure exceeds the cost and complexity of a conventional database, and anything where reversibility of errors is a requirement rather than a bug. The canonical honest assessment remains that most proposed blockchain applications would work as well or better on a conventional database — and the ones that would not are genuinely interesting, and fewer than fifteen years of investment has produced.
Tomorrow we are going somewhere with considerably more physical substance and considerably less volatility — advanced manufacturing and the industrial technology revolution. 3D printing at scale, computer-controlled fabrication, digital-physical integration in production, and why the next wave of manufacturing technology may reshape global trade patterns as significantly as the first industrial revolution did. See you then.
Switched On is a daily technology series covering the ideas, systems, and arguments shaping the digital world. Opinionated. Witty. Occasionally wrong. Always worth the argument.
