Policy writer Yaël Ossowski on how the 1970 Bank Secrecy Act quietly became a mass financial-surveillance regime that catches almost no criminals, why the reporting threshold has never been indexed, and how self-custodied Bitcoin is the non-violent way out.
↓ Jump to the video and timestamps
Yaël Ossowski opened the conversation with the part of the story that should bother everyone and somehow doesn't. The Bank Secrecy Act, the law that turns every bank into a reporting arm of the government, didn't get weaponized against cartels or terror financiers. It got pointed at ordinary people.
"The Bank Secrecy Act was actually weaponized against ordinary people. People who bought guns, people who went to Trump rallies. They were all thrown into this database and then at some point were denied financial services. Not because the government forced these banks to do so, but it's because the banks didn't want to deal with the additional compliance."
That is the whole machine in four sentences. A 1970 law tells banks to watch their customers and report them. The banks, not wanting the liability or the cost, find it easier to drop anyone who trips a flag than to defend keeping them. The government never has to order a single account closed. It just builds the incentive and lets the private sector do the censoring. Ossowski is a fellow at the Bitcoin Policy Institute and deputy director of the Consumer Choice Center, and he has spent a decade in the rooms where this stuff gets written. His framing flips the name on its head: the Bank Secrecy Act has nothing to do with secrecy, and the most striking thing about it is how badly it does the one job it claims to do.
This is the regulatory bedrock underneath everything in the piece on how the digital dollar arrived without a CBDC. The synthetic CBDC is frightening precisely because it gets plugged into the machine Ossowski describes here, a surveillance framework that was already running and was never working.
Ossowski's first move is to strip the euphemism. The Bank Secrecy Act, signed in 1970, sounds like a privacy protection and is the opposite of one. It compels banks to report customer activity to the government, funneling it to the Financial Crimes Enforcement Network. Two instruments do the work, and they get conflated constantly: a Currency Transaction Report fires automatically on cash transactions over $10,000, no suspicion required, while a Suspicious Activity Report is filed at a bank's discretion when it decides something looks off. Once a report is filed on you, Ossowski's description is that it functions like a quiet notch on your record, the kind of internal flag that makes institutions reluctant to deal with you without your ever being told it exists.
What surprises people, he said, is who hates this most. It isn't only the privacy crowd.
"It has nothing to do with secrecy. And realistically it's just a bank regulation. And you'd be surprised to learn that most banks actually hate this. They actually don't like it. They don't like complying with it. It's very burdensome. They need to hire entire departments that only do compliance related to it."
He has watched the size of that industry up close. At European Bitcoin and blockchain conferences, he noted, something like three-quarters of the sponsors are compliance and identity-verification companies, the vendors that exist only because the rules force every exchange to buy chain-surveillance and KYC tooling. The compliance burden didn't shrink crime. It grew a sector.
The single detail that exposes the whole regime is the number. The $10,000 reporting threshold was set in the early 1970s and has never been indexed to inflation. The Government Accountability Office calculated in 2025 that an inflation-adjusted figure would sit around $72,880, and that indexing it would have cut reporting volume by roughly 90 percent over the past decade. Nobody expanded the law's reach on paper. They simply left the number alone while the dollar lost value around it, and the surveillance net widened year after year by default. Ossowski's point is that this is the quietest possible way to grow a surveillance state: you don't pass anything, you just refuse to fix the math.
The part Ossowski drives hardest is effectiveness. For all the cost and all the surveillance, the system barely accomplishes its stated purpose. The United Nations estimates that less than 1 percent of laundered criminal proceeds, probably around 0.2 percent, is ever seized. Marty pointed to the same gap from the industry side: the money-laundering interdiction rate is rounding-error small, and there's little reason to think the government has gotten meaningfully better at it. Ossowski pointed to Nick Anthony at the Cato Institute, who has gone through the raw figures and found banks filing millions upon millions of suspicious activity reports a year, a volume no human team could ever actually review. The reports get filed, they go into a database, and they sit there until someone is already under investigation, at which point the file becomes ammunition rather than detection.
His point isn't that money laundering is fine. It's that a regime which surveils everyone and stops almost nothing was never really an anti-crime regime. The crime-fighting is the justification. The surveillance is the product.
The clearest sign that the regime expands toward ordinary people, not away from them, is what happened recently at the southern border. Ossowski described how the reporting requirement was ratcheted down so aggressively in border states that money-services businesses had to report transactions as small as a few hundred dollars, sweeping in anyone living in Arizona, Texas, Nevada, or California who happened to send a normal remittance. A threshold built for the appearance of catching kingpins ends up logging a construction worker wiring money to family.
Then there's the data itself, which closes the loop on the harm. The law forces institutions to hoard identity and transaction records, and there is no requirement that the data be held securely, no mandate that it be encrypted. When it leaks, and Ossowski's point is that it constantly does, it hands criminals a map of who holds significant assets and where to find them. He was blunt that this lands hardest on younger holders and minorities who stacked sats through centralized exchanges precisely because they didn't trust the traditional system, and who become physical targets the moment a KYC database spills. A regime sold as safety manufactures a fresh physical-security risk for the people it surveils, Bitcoin holders very much included.
Ossowski's constructive argument is the spine of the episode, and it's the part that separates Bitcoin policy from crypto lobbying. He's wary of prescriptive rules, because detailed regulation tends to build moats that help the largest incumbents and crush the next builder. His preferred model is what he calls "negative freedoms," legislation that restrains the government rather than micromanaging the technology. The template is the First Amendment: "Congress shall make no law." It doesn't grant a right, it forbids the state from intruding on one that already exists.
He grounded it in real examples most people have never connected to financial privacy. In the 1990s, state legislatures started passing preemptions that stripped local governments of the power to do petty things, like Michigan blocking municipalities from banning plastic straws, or Texas reining in Austin's city council. The point of a preemption isn't the straw. It's that constraining a level of government is how you protect a freedom durably. He wants Bitcoin policy written the same way: don't pass a sweeping affirmative "Bitcoin law" that a future administration can repeal on day one, just find every place the government is reaching into Bitcoin usage and tie its hands there. As he put it, bitcoiners aren't asking for a license to do something. They're asking the government to leave an existing right alone.
That principle is why he's skeptical of the crypto industry's affirmative agenda, the push to force government agencies to adopt particular tokens and to write rules that, conveniently, only the well-capitalized can comply with. "The big companies have the resources to fight it," he said. "It's the small developers and people who come up with cool stuff who get harmed." Regulation, in his telling, is how incumbents pull the ladder up.
Two threads from the policy weeds are worth pulling out. The first is a phrase Ossowski found buried in the CLARITY Act: a "mature blockchain system," defined roughly as a network so decentralized that no single person controls more than a 20 percent stake. He'd never heard the term in a single meeting with legislators, which is its own tell about how this language gets drafted, but the practical effect is that it would reclassify Bitcoin exchanges as financial institutions and, he claimed, raise the suspicious-activity reporting threshold on crypto withdrawals from around $2,500 to $10,000. A small improvement bolted onto a much larger and messier bill.
The second is the one with real teeth. Marty laid out the administration's apparent ambition to effectively merge the Treasury and the Federal Reserve, removing the last independent brake on fiscal policy so they can, in his words, "turn things on turbo." Ossowski's response was that this puts the country "one stop to funny money," a president setting interest rates at zero forever because that's what a real-estate guy wants. The wrinkle he kept circling is that Treasury Secretary Scott Bessent is himself a bitcoiner who understands sound money, which makes the whole thing harder to read: the people now positioned to run the printing press are the same people who spent years criticizing it. With the federal debt around $38 trillion, the temptation to capture the Fed and monetize the deficit isn't theoretical. It's the most predictable thing in Washington, and it's the reason an exit that doesn't depend on Washington matters at all.
Asked for his short list, Ossowski didn't reach for a Bitcoin reserve or a grand framework. He wants two specific things. First, a de minimis tax exemption so that everyday Bitcoin spending below a threshold doesn't have to be reported to the IRS, which is the difference between Bitcoin being usable as money and being a taxable event every time you buy a coffee. Second, the Blockchain Regulatory Certainty Act, which would protect developers who write non-custodial code from being prosecuted for what users later do with it, the exact issue at the center of the Samourai and Tornado Cash cases. On the Bank Secrecy Act itself, he pointed to Senator Mike Lee's Saving Privacy Act as the vehicle he'd most like to see move, and to a quietly bipartisan opening: even a Democrat like New Jersey's Andy Kim has started saying in hearings that the reporting regime is hurting ordinary constituents.
His closing point was about who carries the argument. It can't be the Bitcoin lobby, he said. It has to be the person who owns a hair salon in Minneapolis or an auto-parts store in Detroit, the ordinary user who just wants to save in Bitcoin, spend it, and be left alone. Which is, more or less, the entire TFTC thesis stated by someone who does policy for a living.
The Bank Secrecy Act regime Ossowski dissects is the substrate the synthetic CBDC runs on. The full account shows how the GENIUS Act folds private stablecoins into this same machine, requiring issuers to comply with the BSA and maintain the ability to freeze and block transactions, which is why the most surveillable money in history is being built on top of a framework that was already failing.
Yaël Ossowski is a writer and policy advocate, a fellow at the Bitcoin Policy Institute and deputy director of the Consumer Choice Center, focused on financial privacy and consumer-facing technology policy.
