The Second Crypto War: A Private Ethereum
Where have all the cypherpunks gone?
It's become a refrain in crypto, almost a complaint. After years of ETFs, corporate chains, BlackRock stablecoins, and memecoins named after presidential pets, you'd be forgiven for thinking the original vision died somewhere between the Coinbase IPO and the first Bitcoin futures contract. Privacy? That's for criminals. Decentralization? A nice marketing term. Permissionless finance? Sure—as long as you complete KYC with three forms of ID and a retinal scan.
I've felt this cynicism creeping in myself. Watching the industry congratulate itself for "institutional adoption" while the actual technology that matters—the infrastructure of freedom—gets prosecuted out of existence.
Then I spent an hour listening to two physicists explain why they've been grinding for eight years on privacy infrastructure that most of the industry had written off as impossible. And I remembered: the cypherpunks never left. They just kept building.
The Second Crypto War Is Here
Let me be direct about the current landscape.
In May 2024, a Dutch court sentenced Tornado Cash developer Alexey Pertsev to 64 months in prison for money laundering. The judge declared from the bench that "Tornado Cash in its nature and functioning is a tool intended for criminals." Pertsev wrote code. Open-source code. Code that anyone could run. He's now serving four and a half years.
In August 2025, a U.S. jury convicted Roman Storm—another Tornado Cash co-founder—of operating an unlicensed money transmitting business. The jury deadlocked on the more serious charges, but the conviction stands. Storm is now asking the judge to toss the conviction.
The Samourai Wallet founders fared worse. In April 2024, Keonne Rodriguez and William Lonergan Hill were arrested—Rodriguez in Pennsylvania, Hill extradited from Portugal. They pleaded guilty. In November 2025, they were sentenced to four and five years in federal prison for building a Bitcoin mixer.
Let me repeat that: developers are going to prison for writing privacy software.
This is the second crypto war. And unlike the first one, we're losing.
The First Crypto War: A Brief History of Victory
In 1991, a software developer named Phil Zimmermann released a program called Pretty Good Privacy—PGP—that allowed ordinary people to encrypt their messages with military-grade cryptography. The U.S. government classified encryption software as munitions under the International Traffic in Arms Regulations. Zimmermann faced a three-year criminal investigation for exporting weapons. His crime: making software accessible on the internet.
The cypherpunks fought back with subversive brilliance. Zimmermann convinced MIT Press to publish PGP's source code as a book, then ship it to European bookstores. If Phil Zimmermann was an illegal arms dealer, then so was one of America's most prestigious universities. The Electronic Frontier Foundation challenged the Clipper Chip—the NSA's attempt to mandate government backdoors in all encryption. They won.
In 1996, the Justice Department dropped its investigation. Federal courts ruled that encryption is protected by the First Amendment. In 2000, export restrictions collapsed. Strong encryption became legal, then ubiquitous. Today, every HTTPS connection, every secure messaging app, every online bank transaction exists because the cypherpunks won the first crypto war.
Zimmermann later reflected: "PGP ignited the decade of the Crypto Wars, resulting in all the western democracies dropping their restrictions on the use of strong cryptography. It was a storied and thrilling decade, and a triumph of activism for the right to have a private conversation."
That right is now under assault again. And this time, the battleground is blockchain.
Into This Dystopia Walks Aztec
A few weeks ago, at DevConnect in Bangkok, a network called Aztec launched its "ignition chain"—a fully decentralized layer 2 on Ethereum designed from the ground up for privacy. Not privacy as a feature. Privacy as infrastructure.
The founders—Zac Williamson and Joe Andrews—are not the typical crypto operators. Williamson is a former particle physicist. Andrews comes from materials science. They've been building Aztec for eight years. Eight years of grinding on cryptography problems while the rest of crypto chased yield farms, NFT drops, and meme seasons.
When Ryan Adams asked them on the Bankless podcast why they've stuck with it for so long, Williamson's answer was brutally simple:
"What else is there? If I left the space or pivoted into some random piece of junk that I didn't believe in, then what's it all for? This is the opportunity of a lifetime. I wouldn't trade it for anything. It's do or die."
Do or die. That's cypherpunk energy.
The Private World Computer
Here's what makes Aztec different from every privacy solution that came before: it's not just private Bitcoin. It's a private Ethereum.
The distinction matters enormously. Zcash, Tornado Cash, Railgun—these provide private transactions. You can shield a balance, transfer it, unshield it. Useful, but limited. What you can't do is build complex applications with private state, private logic, and private composability between contracts.
Aztec does exactly that. It's a fully programmable blockchain where every transaction is a zero-knowledge proof by default. Smart contracts can have private and public components. They can call other contracts. They can compose atomically. The user generates proofs locally on their device—the sensitive data never leaves their phone.
Williamson explained what this enables:
"Let's say you're building a DBank. The entry point is your account contract with account abstraction—maybe you're signing in with your Google account. Your transaction gets forwarded to a darkpool contract, which triggers trades between your USDC and some real-world asset token. That asset contract calls out to an identity contract to verify you're allowed to hold it. Everything happens atomically in a single transaction. And an observer looking at the blockchain? They don't know who you are, the contracts you're interacting with, or the data you've modified. They see nothing."
This is the private world computer. Not hiding from the blockchain—building on it with full privacy.
The ZK Passport Breakthrough
One of the persistent criticisms of privacy in crypto is the compliance problem. If everything is private, how do you distinguish good actors from bad? How do you build applications that need some understanding of identity without recreating the KYC surveillance apparatus?
Aztec's answer is something called ZK Passport—a project they funded that uses the NFC chips already embedded in modern passports. Here's how it works:
Your passport has a cryptographic chip that can be read by your phone. That chip contains data signed by your government—your photo, nationality, date of birth. ZK Passport reads that chip, then generates a zero-knowledge proof that attests to specific claims without revealing the underlying data. You can prove you're not on a sanctions list without revealing your name. Prove you're over 18 without revealing your birthday. Prove you're from a specific country without revealing which one.
The data never leaves your phone. The proof goes to the blockchain. Aztec has convinced Swiss regulators that this constitutes a valid sanctions check—meaning you can participate in their token sale using ZK Passport instead of traditional KYC.
This matters beyond Aztec because AI is about to break traditional identity verification entirely. Google Gemini can already generate deepfakes good enough to fool selfie-based KYC systems. Within months, the entire "take a photo with your ID" paradigm will be worthless. The only identity verification that will survive is cryptographic—proving things about credentials without revealing them.
Aztec is building that future now.
"I'm Not Worried They Won't Understand. I'm Worried They Do."
Given what's happening to other privacy developers, I asked the obvious question: aren't they afraid?
Williamson's response was illuminating:
"I'm not worried that regulators won't understand what we're building and come after us. I'm worried that regulators do understand what we're building and come after us anyway. Because the last few years have shown us that regulators can often act in extraordinarily bad faith. You have the de jure mandate—protect consumers—and then you have the de facto mandate: protect incumbent economic and financial elites."
He continued:
"If Aztec succeeds, it creates a network where the barriers to entry for providing financial services crush through the floor. You no longer need giant information moats, enormous pools of capital, access to elite networks of privilege. You can build decentralized financial systems that compete directly with centralized versions. If we're successful, they're not going to like it. And they're going to dress up their concerns in the language of privacy and safety when what they're really upset about is losing money."
This is the real battle. Not criminals versus law enforcement—incumbents versus disruption. The first crypto war was about whether strong encryption would be legal. The second crypto war is about whether financial privacy will be criminal.
Fully Decentralized From Day One
Most layer 2s follow a "progressive decentralization" roadmap that somehow never reaches its destination. Launch with a centralized sequencer. Promise decentralization later. Keep promising.
Aztec launched fully decentralized on day one. No centralized sequencer. Permissionless validator set—currently about 600 nodes. Decentralized proving. Community governance from the start.
Andrews explained why:
"Privacy shouldn't be owned by anyone. It's not owned by anyone on the internet—we take that for granted. When we use SSL transactions, this podcast is encrypted. We want to make sure that when this amazing privacy comes to Ethereum, it's not controlled by one entity or team."
The incentive structure matters here. Other layer 2s resist decentralization because their centralized sequencers capture all the fees. Aztec has different economics—their value proposition is privacy, not speed or cost. Decentralization isn't a cost to minimize; it's the core product.
This also provides legal protection. Aztec isn't a service that processes transactions—it's a protocol that anyone can run. The same distinction that ultimately protected Bitcoin itself, and that the courts recognized when overturning OFAC's sanctions on Tornado Cash's autonomous contracts.
The Privacy Layer for All of Ethereum
Here's the part that surprised me most: Aztec isn't trying to rebuild DeFi from scratch. They're building a privacy layer that blankets existing Ethereum.
Through what they call "private intents," users can route transactions through Aztec to access any DeFi protocol on any layer 2 or the Ethereum mainnet—privately. You don't need to bridge your entire portfolio to Aztec. You pass through, execute your transaction with privacy, and exit to whatever destination you choose.
Andrews used a simple example: swapping ETH for USDC on Base.
"You send a transaction on Aztec that says 'I want to swap 10 ETH for USDC on Base.' A relay picks up that intent, fills it, and returns your funds. The world sees 'someone on Aztec wants to swap'—but not who. You get the liquidity of Base, the privacy of Aztec, and your assets end up wherever you want them."
This solves the cold-start problem that kills most new chains. Aztec doesn't need to bootstrap liquidity—it taps into existing liquidity everywhere. It doesn't need to convince protocols to redeploy—it composes with them directly.
The Long Grind Ahead
The ignition chain is live but limited. The gas limit is set to zero—no user transactions yet. Full alpha mainnet comes after audits complete in February 2026. The founders are explicit that this is still early, still dangerous, still experimental.
But the architecture is in place. The validators are running. The proving is decentralized. The governance is community-controlled.
This is what building for the long term looks like. Not shipping fast and breaking things—shipping carefully because the stakes are too high to get wrong. Not chasing the current meta—building infrastructure that will matter in ten years.
When Adams asked what keeps them going after nearly a decade, Andrews put it simply:
"If we're going to onboard the unbanked to blockchain rails where everyone can see every transaction, we've built something worse than the current system. An MEV bot that can predict your next move because it's seen your entire history. I feel compelled to prevent that future. That's what keeps me going."
The Cypherpunks Never Left
The first crypto war taught us that privacy is worth fighting for, that governments will overreach, and that determined technologists can win against seemingly impossible odds. Phil Zimmermann faced years of prosecution and emerged victorious. His code became the foundation of internet security.
The second crypto war is uglier. Pertsev is in prison. The Samourai founders are serving time. Storm awaits sentencing. The chilling effect is real—developers are afraid to build privacy tools, projects are self-censoring, and the industry has largely capitulated to surveillance-first thinking.
But not everyone.
Aztec represents something increasingly rare in crypto: a project that's building what matters instead of what's easy, that's optimizing for long-term impact instead of short-term extraction, that's willing to absorb regulatory risk because the mission is worth it.
"Do or die," Williamson said. That's not bravado. That's conviction.
The cypherpunks never left. They just kept building. And now, eight years later, they're ready to show what they've made.
Disclaimer: I've created a ZKPassport account and am participating in the Aztec public token sale that is currently underway. The views expressed are entirely my own.
For more on Aztec, visit aztec.network. The full Bankless interview is worth your time.
Related Reading: AI and the War Machine — How centralized AI labs abandoned their humanitarian missions to build weapons systems for American empire.