The series that translates national-security and AI-policy arguments across partisan lines, because the stakes are too high for tribal shorthand.
Three Bills, One Fight
What Happened, In Order
On Wednesday, April 29, the U.S. House passed a 3-year extension of FISA Section 702 by a vote of 235 to 191. The vote came twelve days after the original five-year reauthorization failed on the House floor on April 17, after which Congress passed a clean 10-day extension to buy negotiating time. That extension expires at midnight tonight. As I write this on Thursday morning, the Senate has roughly fifteen hours to either pass the House bill, send it back with amendments, or let Section 702 lapse.
The bill the House sent the Senate is a hybrid. The first half is the most substantive surveillance-reform legislation Congress has produced in two decades: a warrant requirement for U.S. person queries of 702-collected data, a written justification submitted to the Office of the Director of National Intelligence for every targeted query, criminal penalties of up to five years in prison for misuse of the database, and a mandatory monthly civil-liberties review. These are the reforms that civil-liberties hawks and intelligence reformers have demanded since the 2008 enactment of the FISA Amendments Act. They were negotiated in good faith over multiple years, across two administrations, and survived a House Intelligence Committee that historically resists them.
The second half of the bill is a permanent statutory ban on the Federal Reserve issuing a central bank digital currency. It is not surveillance reform. It is not even surveillance-adjacent, except in the loose sense that any payment system creates surveillance potential. It was added by House leadership at the last hour to win conservative holdouts who were not going to vote for the surveillance reform alone. Senate Majority Leader John Thune has now said publicly that the CBDC ban is "not happening" in tandem with FISA 702 reauthorization and that the House bill as drafted is "dead on arrival" in the Senate. Senate Democrats, who view a CBDC ban as a separate fight they are not willing to lose, are aligned with him on the procedural objection even where they disagree on the underlying CBDC question.
Here's What You Need to Know in 30 Seconds
If the Senate strips the CBDC language and passes the rest, the bill goes back to the House, where Speaker Mike Johnson will have to decide whether to bring it to the floor before midnight without the holdouts' price tag. If the Senate passes the bill as-is with the CBDC ban, Senate Democrats will filibuster, the cloture vote fails, and 702 lapses anyway. If the Senate does nothing, 702 lapses at midnight. There is no clean win available before the deadline. Whichever path the Senate picks, somebody is choosing between letting a critical foreign-intelligence authority go dark for an indeterminate period and accepting a procedurally aggressive rider on unrelated policy. The reforms that actually matter, the ones civil-liberties advocates have been demanding for fifteen years, are now leverage chips in a fight over a Federal Reserve policy that nobody in the surveillance-reform negotiation ever discussed.
The Hawk Case: 702 Has to Pass, and the CBDC Ban Is Not Crazy
The hawk position starts from operational necessity. Section 702 is the legal authority that lets the Intelligence Community collect foreign intelligence on non-U.S. persons reasonably believed to be located outside the United States. The IC has documented, in unclassified summaries to Congress, that 702 is the source of a substantial fraction of the President's Daily Brief and the largest single contributor to the IC's terrorism, counterproliferation, and cyber-threat reporting. A 702 lapse means real loss of capability, even if it is temporary. Past short-term lapses have been measured in hours and have caused detectable disruption. A multi-day lapse, which is what happens if the Senate sends the bill back to the House and the House cannot reconvene quickly enough, would be operational malpractice.
Given that, the hawk says, the choice the Senate faces is not between a clean reauth and a poison-pilled reauth. The choice is between a poison-pilled reauth that includes the most substantive 702 reforms in decades and a lapse. Clean reauth is not on the table. It was never on the table after April 17. The House could not pass a clean reauth, and the Speaker spent two weeks trying. The bill that came out of the House is the bill that exists. Senate Democrats refusing to accept it on the basis of a CBDC ban they were going to oppose anyway is, in the hawk's framing, performative. They get to vote against the CBDC ban in a future fight. They do not get to revive a clean 702 reauth that died on April 17.
The hawk also makes a substantive argument about the CBDC ban itself. A Federal Reserve central bank digital currency would, if implemented, create the most comprehensive financial-surveillance infrastructure in U.S. history. Every transaction, every counterparty, every timestamp, in a single database accessible to the federal government without intermediation by a private bank. That is precisely the kind of capability that the civil-liberties reformers in the 702 fight have spent fifteen years arguing should not exist. A statutory ban on the Fed issuing a CBDC is, on its merits, a civil-liberties win. The fact that it is in the wrong bill is a procedural complaint, not a substantive one. If you are a civil-liberties reformer who supports the 702 safeguards, you should also support a permanent CBDC ban.
The hawk's bottom line: Pass the bill. The 702 reforms are real. The CBDC ban is defensible. The cost of a lapse is unacceptable. Senate Democrats who let 702 die over the CBDC question will own that outcome, and the political ad writes itself.
The Reformer Case: The Reforms Were Real Until They Got Used As Leverage
The reformer position agrees with most of the hawk's facts and reaches the opposite conclusion. The 702 reforms in the House bill are, in fact, the most substantive surveillance reforms in two decades. The reformer agrees that warrant-for-query is a major win. The reformer agrees that the ODNI written-justification log is a real audit trail. The reformer agrees that criminal penalties for misuse impose individual accountability that has never existed in this regime.
What the reformer rejects is the framing that the only choice available is the bill in front of the Senate. The reason that is the only choice is that the Speaker decided to attach a CBDC ban to the bill. The Speaker had options. He could have brought the surveillance-reform bill to the floor without the CBDC language and dared his conservative holdouts to vote against substantive 702 reforms with criminal penalties for IC misuse. The conservative holdouts, almost all of whom have spent the past several years complaining about FBI abuse of 702, would have had to choose between voting for reforms that target exactly that abuse or voting against them. The Speaker did not run that play. He chose the path that maximized his caucus management problem and minimized the substantive policy bargain. That choice is not a force of nature. It was a choice.
The deeper reformer concern is that once must-pass leverage gets used to ride non-germane policy through, every future reauthorization works the same way. Section 702 comes up for reauthorization roughly every four to five years. The next reauth, around 2029, will face exactly this dynamic. Whichever party holds the Speakership will have an incentive to attach whatever unrelated policy fight they need votes on to the reauth bill. The other party will have an incentive to filibuster. The IC will threaten lapse. The reform negotiation will be weighed against the rider. Some bargains will be made. Some reforms will be abandoned. The net effect is that the substantive policy work — the part where reformers and intelligence professionals figure out what the surveillance regime should actually look like — gets repeatedly devalued by the procedural game.
On the substantive merits of the CBDC ban, the reformer is split. Some civil-liberties reformers do support a permanent CBDC ban on exactly the surveillance grounds the hawk identifies. Others view the CBDC question as premature, since no Federal Reserve CBDC is currently in development and the Fed's own studies have been cautiously skeptical. The split does not matter for purposes of the immediate fight, because the question on the floor is not about the merits of a CBDC ban. The question is about whether unrelated policy belongs in must-pass surveillance legislation. The answer to that question, the reformer says, has to be no. Otherwise the surveillance-reform negotiation is just a delivery vehicle for whatever else is on the table that week.
The reformer's bottom line: Strip the CBDC ban. Pass the surveillance reforms. If the House refuses to concur, that exposure belongs to the Speaker, who chose the leverage play. The reform bargain that took five years to negotiate cannot be allowed to become a rider chip every reauth cycle. The principle is bigger than this one bill.
Where They Actually Agree
Strip the rhetoric and the hawk and the reformer agree on more than either side will say in public.
Both agree the surveillance-reform safeguards in the bill are substantive. Both agree the CBDC ban does not belong in a surveillance bill. Both agree that letting 702 lapse outright would be operationally damaging and politically expensive. Both agree that whatever happens tonight, the same dynamic will repeat at the next reauth. Both agree that the structural problem is not Section 702 specifically. The structural problem is that every must-pass authorization in this Congress becomes a Christmas tree, because that is the only place where leverage exists.
Both also agree, privately, that the Senate's path is being chosen for them by the calendar. The 10-day extension expiring tonight was itself a procedural compromise that bought time without solving anything. The deadline is a manufactured forcing function. Everyone in the room knows that another extension could be passed in twenty minutes if leadership wanted to. The fact that nobody is proposing one is itself a tactical choice, not a constraint. Both sides are using the deadline as leverage. The Speaker is using it to get the CBDC ban through. The Majority Leader is using it to force a Senate vote on a House-drafted bill rather than negotiating amendments. The intelligence agencies are using it to remind everyone what is at stake. The civil-liberties reformers are using it to demand the safeguards. Nobody is innocent of using the clock.
Where They Don't (And Shouldn't Pretend To)
On whether the CBDC ban is good policy on its own merits. The hawk argues it is a civil-liberties win that any reformer should support. The reformer argues it is, at best, a premature answer to a hypothetical capability and, at worst, a sop to a constituency that conflates monetary policy with surveillance policy. Both positions have defenders inside the broader civil-liberties coalition. The disagreement is real and not bridgeable in the next fifteen hours.
On who owns the political cost of a 702 lapse. The hawk argues Senate Democrats own the lapse if they refuse to swallow the rider. The reformer argues the Speaker owns the lapse if he refuses to bring a clean bill. Each side is partially right. The lapse does not have a single author. It has a Speaker who attached a poison pill, a Majority Leader who did not negotiate, and a Senate caucus that did not agree to a deal it views as procedurally bad-faith. Apportioning the political cost is a question for whoever writes the campaign ads in the fall.
On what to do next time. The hawk's prescription is to build a structural deal in advance of the next reauth so that the dynamic does not repeat. The reformer's prescription is to separate substantive surveillance reform from reauthorization deadlines entirely, by passing the reform pieces through regular order whenever they are ready and letting the reauth itself be a clean reauth. Neither prescription has been attempted. The hawk's would require leadership to negotiate further in advance than it has shown any willingness to. The reformer's would require Congress to legislate surveillance reform without a forcing function, which the existing record suggests it cannot do.
Here's My Two Cents
The substance of this bill is, on its own, one of the better pieces of surveillance-reform legislation Congress has produced this decade. Warrant-for-query for U.S. person searches is a real win. The ODNI written-justification log is a real audit trail. Criminal penalties for misuse are real individual accountability. These are the asks that civil-liberties advocates have been making since 2008. They are inside the bill. They are negotiated. They have a House majority. The fact that I have to write that sentence in past tense in another twenty-four hours, depending on how the Senate vote goes, is the structural failure I want to underline.
The procedural posture of this bill is also one of the worst examples this decade of must-pass leverage being used to ride unrelated policy through a vehicle the other party cannot afford to kill. The CBDC ban does not belong in a surveillance bill. It is there because the Speaker counted votes and that was the price of his conservative holdouts. The holdouts demanded it. The Speaker delivered it. The bill is now a hostage to a separate fight that has nothing to do with the surveillance regime its first half is meant to reform.
The reason this matters beyond tonight is that the same dynamic is going to repeat. The next must-pass authorization, whether it is the NDAA, an appropriations cycle, or the next reauth of some other authority, will face exactly the same caucus-management math. The Speaker, whoever it is, will count votes and conclude that the only way to pass the must-pass bill is to attach the off-topic policy that gets him the holdouts. The other party, whichever party that is, will have to choose between accepting the rider or letting the must-pass authority lapse. The substantive bargain inside the bill — the surveillance reform, the defense authorization, the appropriations specifics — will be devalued by the procedural game. That is what we have done to legislation in this Congress. We have made every must-pass bill a delivery vehicle for whichever rider the holdouts demand. The reform negotiations that take years to build are now bargaining chips for unrelated fights.
What strikes me, watching this play out, is that the people who actually care about surveillance reform — the IC reformers who built the warrant-for-query language, the staff lawyers who wrote the ODNI logs, the IGs who pushed for criminal penalties — are not the people whose policy preferences are being expressed by the procedural game. The CBDC ban was negotiated by a different set of members entirely, with different policy priorities, in a different room. The surveillance-reform negotiation produced a bipartisan bargain. The procedural game subordinated that bargain to a different bargain made among a different set of people. The reformers are the ones whose work is now hostage. They had no leverage in the procedural fight, because surveillance reform is not a constituency that delivers floor votes the way conservative monetary-policy holdouts do.
That asymmetry is the structural problem, and it is bigger than 702. Constituencies that deliver floor votes write riders. Constituencies that build careful bipartisan policy bargains do not. So the procedural game systematically favors the former over the latter. The longer-term consequence is that careful bipartisan policy work decays as a category, because the people doing it cannot protect their work from being repurposed at the procedural-leverage stage. We end up with fewer substantive bargains and more hostage-takings.
If I were advising the next surveillance-reform negotiation team, I would tell them to do something this team did not do. Do not wait for reauthorization. Pass the substantive reforms as standalone legislation, even if it takes a separate vote and even if it has to fail two or three times before it builds the coalition. Make warrant-for-query a standalone bill. Make the ODNI logs a standalone bill. Make criminal penalties for IC misuse a standalone bill. Each of those gets a clean up-or-down on its own merits. The reauthorization itself, when it comes, is then a clean reauth, no riders, because the riders have been pre-empted by the standalone reforms already in law. That breaks the leverage cycle for surveillance specifically. It costs years. It is worth years.
But that requires Congress to legislate substantive policy without a forcing function. Which Congress has not done in twenty years. Which is the structural problem this bill, and the next one, and the one after that, are not going to solve.
The Senate vote tonight will reveal which is more important to the chamber: the surveillance reform or the CBDC question. Either answer is interpretable. If the Senate strips the CBDC ban and passes, the message to the next leverage play is that the Senate will not accept non-germane riders even on must-pass bills. If the Senate accepts the CBDC ban and passes, the message is that the riders work. If the Senate lets 702 lapse, the message is that nobody owns the cost cleanly enough for the leverage to be decisive. Each outcome has a follow-on. Each follow-on shapes the next reauth. None of them gets us back to the substantive policy bargain that took five years to negotiate.
The reforms are real. The leverage is real. Both can be true. Both are, tonight.
Sources
This briefing draws on the following reporting and primary documents. The links are direct to the original publications.
- House advances 3-year extension of FISA 702 surveillance program — NPR, April 29, 2026.
- House passes FISA Section 702 renewal as Senate faces midnight deadline — Fox News, April 29, 2026.
- House passes 3-year FISA 702 extension — Nextgov/FCW, April 29, 2026.
- House approves reauthorization of FISA 702 warrantless spy powers — The Hill, April 29, 2026.
- House passes FISA extension ahead of Thursday deadline — CBS News, April 29, 2026.
- US Congress extends controversial surveillance power under FISA for 10 days — Al Jazeera, April 17, 2026.
- Congress approves brief extension of FISA surveillance; disputes unresolved — Washington Times, April 17, 2026.
- Three things to know about FISA Section 702: Congress passes short-term extension — CNBC, April 17, 2026.
- Section 702 of the Foreign Intelligence Surveillance Act (FISA): 2026 Resource Page — Brennan Center for Justice.
- FISA Section 702 and the 2024 Reforming Intelligence and Securing America Act (RISAA) — Congressional Research Service, Library of Congress.
Related Briefings
Anna R. Dudley writes on national security, intelligence policy, and the places where hawks and reformers need to find each other. Bipartisan Translation is the weekly series for the conversation that is not happening on cable news. Subscribe at annardudley.substack.com.