The United States economy is facing several systemic vulnerabilities that are beginning to converge in a nightmare scenario. As the broader economy slows, America’s Gross Domestic Product (GDP) growth is being upheld almost exclusively by massive investments in Artificial Intelligence (AI) technologies. However, the boom in AI investment is itself concerning, as valuations and expectations of future growth become increasingly detached from revenue realities. Many companies are funding their AI investments through complex, esoteric financial vehicles that have become deeply intertwined with the banking and financial system. All of this is occurring alongside a slowing economy, a weakening dollar, and a federal fiscal situation degraded by decades of unproductive borrowing. If the current wave of AI investments is indeed a bubble — as even many AI executives increasingly believe — then it has the potential to undermine the entire American economy.

If this bubble does pop and cascade throughout the economy, policymakers will likely fall back on their traditional model of crisis resolution: privatizing gains while socializing losses by bailing out the big banks and corporations. This approach was cemented during the 2008 Global Financial Crisis, when the federal government spent $250 billion bailing out major banks. During the — much more contained — bank failures of 2023, the federal government again bailed out Silicon Valley Bank, Signature Bank, and First Republic Bank.¹ Under this dynamic, the federal government has established an implicit promise to corporations and financial institutions that they will use taxpayer dollars to rescue them from the consequences of their own decisions.

The United States cannot afford another bailout. For decades, the United States has squandered its unique fiscal advantage on corporate rescues, wars of choice, and tax cuts that produced no lasting productive capacity. The well is not infinite. The dollar has fallen nearly 11% in 2025²; Treasury auction demand is softening³; and the world's patience with American fiscal irresponsibility is wearing thin⁴. The borrowing capacity we have left must be preserved for genuine public investment—infrastructure, industrial policy, climate adaptation—not for rescuing speculators from the consequences of their bets.

The AI sector is approaching a potential correction. If the bubble bursts, it won't stay contained — any losses will cascade into the banking system, triggering insolvencies and mass layoffs in a self-reinforcing spiral. The need for a new framework for how to deal with these crises has never been more urgent.

This report proposes a new framework for managing future financial crises called the Won't Get Fooled Again Act (WGFAA). The WGFAA is designed to update the mechanisms of financial crisis resolution in two ways.

First, it mandates that failed systemically important institutions be converted into mission-driven public utilities rather than bailed out or liquidated. The process of converting failed institutions will be funded entirely by industry levies, rather than taxpayer dollars.

Second, it expands the definition of systemic importance to encompass the flow of computation, data, and intelligence. Just as the collapse of a major bank threatens the credit lifeline of the real economy, the disorderly failure of a major AI provider or compute provider threatens the digital infrastructure upon which national security, economic competitiveness, and public services increasingly rely. We call these institutions "Systemically Important Technology Institutions" (SITIs).

The framework applies separately to banking and technology. Failed banks become community-focused public financial institutions, modeled on Germany's Sparkassen system and the Bank of North Dakota. Failed AI providers become public research utilities, ensuring that critical computational infrastructure remains operational and that valuable intellectual property doesn't flow to foreign adversaries or domestic monopolists in a fire sale.

Crucially, this proposal rejects purchasing equity at bubble prices. It mandates a strict Asset-Based Valuation Standard for any government acquisition, ensuring taxpayers pay only for tangible infrastructure and proven intellectual property—not speculative goodwill or inflated growth projections. Speculators are wiped out while the public gets lasting assets.

The WGFAA demonstrates that converting failed institutions into public utilities is not merely an ideological preference but a strategic and fiscal necessity—the only path that prevents further concentration of power, preserves America's borrowing capacity for productive investment, and ensures that the transformative power of AI serves the broad American public.

1. Dual Bubbles and the Threat to the American Economy

1.1. Convergence of Financial and Technological Fragility

To understand the necessity of the Won't Get Fooled Again Act, one must first dissect the intricate mechanics of the current economic environment. We are witnessing a symbiotic fragility where the banking sector's health is increasingly predicated on the continued inflation of AI asset values, while the AI sector's liquidity is dependent on a banking system exposed to its high-risk debt.

1.1.1. The AI Valuation Paradox: Capital Expenditure vs. Revenue Reality

The current trajectory of the artificial intelligence sector exhibits the classic hallmarks of a speculative mania, distinct in its capital intensity and reliance on projected rather than realized utility. While the technological promise of generative AI is substantial, the financial valuations of companies in this sector have detached from fundamental economic metrics, creating a "Capex-Revenue Gap" that threatens to destabilize the broader market.

1.1.2. The "Round-Tripping" Revenue Mirage

A significant portion of the revenue growth reported by AI startups is arguably artificial, driven by a phenomenon known as "round-tripping" or circular revenue generation. The interconnections are dizzying: OpenAI has taken a 10% stake in AMD; Nvidia is investing $100 billion in OpenAI; Microsoft is a major shareholder in OpenAI but also a major customer of CoreWeave, in which Nvidia holds a significant equity stake; and Microsoft accounts for almost 20% of Nvidia's revenue on an annualized basis.⁵

Major cloud providers and tech giants invest billions in AI startups, often with the stipulation that the startup spends that capital on the investor's own cloud computing services.⁶ This creates a "trillion-dollar loop" where investment cash on one side of the ledger is transformed into "revenue" on the other, inflating sales figures without representing organic end-user demand. The $100 billion Nvidia-OpenAI deal exemplifies this structure: Nvidia pumps capital into OpenAI to bankroll data centers, and OpenAI fills those facilities with Nvidia's chips. Nvidia is essentially subsidizing one of its biggest customers, artificially inflating actual demand for AI.

The absurdity of current valuations is captured by recent funding rounds. Thinking Machines, an AI startup helmed by former OpenAI executive Mira Murati, raised the largest seed round in history: $2 billion in funding at a $10 billion valuation. The company has not released a product and has refused to tell investors what they are even trying to build.⁷ "It was the most absurd pitch meeting," one investor reported. "She was like, 'So we're doing an AI company with the best AI people, but we can't answer any questions.'"

1.1.3. The Divergence of Investment and Return

Further evidence of the AI market’s core instability can be found in the widening disparity between capital expenditure and realized revenue. Major technology firms and venture-backed startups are investing hundreds of billions of dollars in NVIDIA GPUs, specialized data centers, and energy infrastructure. AI-related capital expenditures have surpassed the U.S. consumer as a primary driver of economic growth, accounting for 1.1% of GDP growth in the first half of 2025.⁸ However, the revenue generation from these investments remains dangerously low.

In the first four months of 2025 alone, global AI startups raised over $50 billion, representing a massive concentration of venture capital into a single sector. Pitchbook reports that nearly two-thirds of deal value in the U.S. went to AI and Machine Learning startups in the first half of 2025, up from 23% in 2023.⁹ This influx of capital has driven valuations to unsustainable levels, with the median revenue multiple for AI companies standing at approximately 30x, and many startups valued at 50x to 100x their actual earnings.

Despite these massive capital inflows, the profitability remains elusive. OpenAI's ChatGPT, by far the most successful generative AI product to date, requires so much expensive computing power to run that it loses money almost every time someone uses it. In the first quarter of 2025, OpenAI reported a net loss of $12.5 billion.¹⁰¹¹ The flagship product of the AI revolution is bleeding $27 billion annually.

The continuing investor confidence in the face of these losses suggests that current valuation models are pricing in future popularity and productivity gains of AI tools. However, these gains have been slow to arrive, even in firms that have embraced the new technology. Research from MIT has found that a staggering 95% of attempts to incorporate generative AI into business operations are failing to deliver increased revenue or productivity. The results cast doubt on whether the wide array of corporate AI customers the markets expect will actually materialize.

Furthermore, the sector faces the challenge of diminishing returns on model performance relative to cost. The cost of training frontier models is growing exponentially—ChatGPT-4 reportedly cost roughly $100 million to train, while successors and competitors are estimating costs in the billions.¹² However, the marginal utility improvements for average users are diminishing, creating a "performance plateau." Users are not willing to pay exponentially higher prices for incrementally better chatbots, yet the infrastructure costs to serve them continue to rise.

An increasing number of market experts have begun warning of the growing instability and uncertain future of the AI sector. Bain & Co.'s annual global technology report says the AI industry will need $2 trillion in annual revenue by 2030 to continue at its current rate¹³. Analyzing similar data, a recent Deutsche Bank report to clients declared such levels of investment as “highly unlikely.”¹⁴ Echoing these concerns, A Bank of England report from October warns that the market could soon experience a "sharp correction" due to overvaluation.¹⁵

1.1.4. The Hardware Choke and Inventory Glut

One significant factor driving both computing costs and company valuations to record highs is the scarcity of computer chips. NVIDIA's market capitalization exceeded $5 trillion in October 2025, rising higher than the GDP of every country except the US and China.¹⁶¹⁷ A significant percentage of its data center revenue is tied to a small handful of hyperscalers and AI cloud providers. As discussed above, this concentration risk is exacerbated by vendor financing, where chip manufacturers or cloud providers invest in the very startups that are buying their chips, creating a circular revenue model that inflates reported sales without generating net new economic value.

These tactics create significant potential for an inventory glut. As companies race to build GPU clusters to avoid being left behind, they risk creating massive overcapacity. If the demand for AI services does not match the massive supply of compute coming online, the rental price of GPUs could crash. In addition to being dire for giants like NVIDIA, this would destroy the business models of "compute-as-a-service" startups, as their projected margins are based on continued scarcity pricing of chips.

1.2. The Banking Sector's "Hidden" Exposure

The risks and inherent instability discussed above is all the more troubling because of the ways our entire financial system is becoming increasingly entwined with the fate of the AI market. The banking sector's exposure to the risks of the AI bubble is multifaceted and opaque, extending far beyond direct commercial loans. It includes exposure to the "shadow banking" system, complex off-balance-sheet arrangements that echo the financial engineering of past crises, and venture debt.

1.2.1. The Shadow Banking Transmission Mechanism

Much of the lending to risky AI ventures has migrated to the "shadow banking" or private credit sector, which is now worth over $3 trillion in assets.¹⁸ Unlike traditional banks, which use customer deposits to back their loans, private credit firms raise the capital for their loans from private investors such as pensions, wealth funds, or high-net-worth individuals. Free of the regulations imposed on traditional banks after the 2008 financial crisis, these companies are free to provide higher-risk loans to both individuals and massive corporations. In theory, private investors bear the losses if these loans default, insulating the wider economy from the kind of market volatility that can kick off a financial crisis. But the reality is more interconnected. Traditional banks lend heavily to these private credit funds, creating a chain of exposure that is difficult to quantify but carries grave risks for the American public.

If AI startups begin to default on their private credit obligations—a likely scenario given high "burn rates" and lack of profitability—the losses will flow back to the balance sheets of the major banks that provide leverage to private credit funds.

Recent failures in the private credit market outside the technology sector, such as the collapse of Tricolor Holdings,¹⁹ have already revealed the opacity and weak underwriting standards that characterize this sector. Tricolor, which both sold used cars and provided subprime auto loans to low-income individuals, declared bankruptcy in September 2025. Despite being a private credit institution, the collapse led to substantial losses for their investors in the traditional banking sector, such as JP Morgan Chase and Fifth Third Bank. In response, JPMorgan CEO Jamie Dimon has warned that more "cockroaches" engaging in similarly risky lending will likely emerge from the private credit sector.²⁰ The contagion risk is significant: if collective wisdom determines that private credit dynamics pose a systemic threat, a self-fulfilling liquidity crisis could ensue, freezing credit markets exactly as they did in 2008.

1.2.2. Cooking the Books

Adding to the alarm is the notable off-balance-sheet financial engineering at the largest AI companies. Major technology firms are using complex accounting structures to hide the true scale of their AI infrastructure bets. In one representative deal, Blue Owl Capital took out a loan for $27 billion to build a data center.²¹ That debt is backed by Meta's payments for leasing the facility. Despite holding ultimate financial responsibility, the $27 billion loan never shows up on Meta's balance sheet. If the AI bubble bursts and the data center goes dark, Meta will be on the hook for a multi-billion-dollar payment.²² Some analysts have flagged the structure of the deal as ripe for fraud, as it effectively conceals massive debts and allows Meta to project a false image of financial health.²³ The financial maneuvering implies a deep fragility beneath the surface of big AI, casting doubt on whether the industry's explosive growth is supported by actual economic reality.

1.2.3. Venture Debt and the Collateral Crisis

Even the more straightforward financing structures carry grave risks for the wider economy. As equity markets have tightened, AI startups have increasingly turned to venture debt to extend their runways. AI startups accounted for nearly 25% of all venture debt dollars in 2024, and this figure has continued to rise in 2025.²⁴ This debt is often secured not by cash flows (which are minimal), but by the assets of the company—specifically, their high-value GPU hardware.

This creates a dangerous collateral vulnerability. In a market crash, the value of these GPUs—the primary collateral backing billions in loans—would plummet due to the aforementioned inventory glut. Banks and debt funds would find themselves holding depreciating hardware assets rather than recoverable capital. The liquidation of these assets to recover losses would further depress hardware prices, triggering a deflationary spiral in the tech hardware market that could make additional firms holding similar collateral insolvent.

1.2.4. Data Center Finance and Energy Exposure

Even if investor dollars continue to flow, many experts warn that energy needs will likely constrain the sector’s growth. The physical infrastructure of AI—hyperscale data centers—is financed through complex project finance structures involving large syndicates of banks. Massive loans for projects like Oracle's "Stargate" involve dozens of financial institutions.

However, the energy bottleneck threatens the viability of these projects. The immense power requirements of data centers are colliding with grid limitations, leading to utility rate hikes and the cancellation of other energy projects to feed AI demand. Major cloud computing providers like Oracle and CoreWeave, despite elite clientele and billions in contracts, are struggling to secure the energy infrastructure they need to complete major projects.²⁵

Physical power shortages or consumer backlash could force the abandonment of half-finished infrastructure projects. A data center that cannot get a power connection is a distressed asset with zero revenue potential. Banks holding the construction loans for these stranded assets would face significant write-downs, mirroring the commercial real estate crisis but focused on digital infrastructure.

Communities are beginning to resist. An exurb in Northern Virginia has found itself essentially surrounded by data centers; previously rural farms have sold out, and residents are asking, "Who do I sue?" This is the beginning of a NIMBY phenomenon that could further delay or halt infrastructure build-out.

1.3. System-Wide Risks to the Financial System

As discussed above, these warning signs are all the more troubling because of the central role AI is playing in fueling the growth of the US economy. In keeping with the consolidated nature of the tech industry, compute is dominated by three major cloud providers (Amazon, Microsoft, Google) and a handful of foundational model developers. This creates "single-point-of-failure" risks for the entire economy, including the financial system itself.

Financial institutions increasingly rely on AI for critical operations: fraud detection, credit underwriting, algorithmic trading, customer service, and risk modeling. A failure, bankruptcy, or service disruption of a major AI provider could paralyze banking operations, creating a direct channel for contagion from the tech sector to the financial sector.

Furthermore, the equity concentration of the "Magnificent Seven" tech company stocks (Apple, Microsoft, Amazon, Alphabet, Meta, Nvidia, and Tesla) in pension funds and institutional portfolios means that a bursting of the AI bubble would cause a massive negative wealth effect. In 2025, 30% of the S&P 500 was held by five companies—the largest concentration in 50 years.²⁶ AI-related enterprises accounted for roughly 75% of gains in the American stock market over the year.²⁷ A crash would erode the capital base of the financial system and damage consumer confidence, potentially triggering a recession that would cycle back to cause defaults on traditional bank loans. This is not diversification; it is concentration risk masquerading as a market rally.

1.4. The Labor Displacement Accelerant

The looming bursting of the AI bubble will not end AI-driven labor displacement—it will accelerate it. This counterintuitive dynamic creates a vicious cycle that compounds the economic damage of a financial correction.

The current wave of AI-enabled layoffs is already staggering. In October 2025 alone, employers announced 153,074 job cuts—the highest October total in over 20 years and the highest single month in the fourth quarter since 2008.²⁸^ ^Total layoffs in 2025 have surpassed 1.1 million, crossing the one-million mark faster than any year since the pandemic.²⁹ Technology and warehousing were the hardest hit, with tech companies cutting 33,281 jobs in October (six times September's figure) and warehousing companies announcing 47,878 cuts (a 4,700% month-over-month increase).³⁰

According to the World Economic Forum's 2025 Future of Jobs report, 41% of employers worldwide intend to reduce their workforce in the next five years due to AI automation.³¹ Anthropic CEO Dario Amodei has predicted that generative AI could wipe out up to half of entry-level white-collar jobs. Klarna has shrunk its headcount by about 40%, in part because of AI.³² Salesforce laid off 4,000 customer support roles, stating that AI can do 50% of the work at the company.³³ Duolingo has stopped using contractors for work that AI can handle.³⁴

Here is the critical dynamic: during the current boom, companies have not tried very hard to realize AI productivity gains. Flush with cheap capital and optimistic projections, firms have been content to experiment with AI while maintaining existing headcount.

A financial crash changes this calculus entirely. Under severe cost pressure, companies will finally do the hard work of restructuring operations around AI capabilities. They will discover productivity gains that seemed elusive during the boom—not because the technology has improved, but because desperation focuses the mind. The layoffs we have seen so far are a preview; a crash would trigger a tsunami.

This creates a doom loop:

AI bubble bursts → asset values collapse, credit tightens → Companies face cost pressure → aggressively implement AI to cut headcount → Mass layoffs → consumer spending falls, demand destruction → Recession deepens → more companies face cost pressure → Return to step 2

2. The Fiscal Constraint

2.1. Preserving America's Borrowing Capacity

The Won't Get Fooled Again Act is not merely good policy; it is fiscal necessity. America's ability to borrow—the foundation of its economic flexibility and geopolitical power—has been severely compromised by decades of unproductive spending. We cannot afford another multi-trillion-dollar bailout, and attempting one could trigger a crisis of confidence in the dollar itself.

2.1.1. The True Cost of "Free" Bailouts

For decades, a bipartisan consensus has held that large-scale government borrowing is not merely acceptable but necessary when invested in productive capacity. This view, associated with economists from Keynes to modern proponents of industrial policy, correctly recognizes that a sovereign nation issuing debt in its own currency faces different constraints than a household or business. When borrowed funds flow into infrastructure, education, research, and industry, they generate returns that exceed the cost of servicing the debt.

The tragedy of American fiscal policy is that we have borrowed like believers in public investment while spending like opponents of it.

Since 2000, the United States has added approximately $30 trillion to its national debt. What do we have to show for it?

$4-8 trillion on wars in Iraq and Afghanistan that destabilized the Middle East, created ISIS, and produced no lasting strategic benefit. $1.5+ trillion in tax cuts (2017 Tax Cuts and Jobs Act) that primarily benefited corporations and the wealthy, with no measurable increase in productive investment.³⁵ $400+ billion in direct bailout costs from the 2008 financial crisis, plus trillions more in Federal Reserve facilities supporting financial asset prices.³⁶ Hundreds of billions in implicit guarantees during the 2023 banking turmoil, protecting wealthy depositors and tech firms from the consequences of speculative banking.³⁷ Ongoing annual deficits reaching over 6% of GDP and projected to rise, with no plan for productive deployment.³⁸

What we did not build: A modern passenger rail network (China built 25,000 miles of high-speed rail; we built zero)³⁹. A resilient electrical grid capable of supporting clean energy and advanced manufacturing. Universal high-speed broadband connecting rural and urban America. Domestic semiconductor capacity at scale (until the recent, inadequate CHIPS Act). Public AI research infrastructure ensuring American competitiveness for decades. Affordable housing at scale addressing the affordability crisis.

Instead, we have been drawing down our most valuable national asset—the world's trust in American debt—to finance empty consumption, crisis management, and the socialization of private losses.

2.1.2. The Dollar's Eroding Foundation

While this bipartisan consensus still largely controls Washington, the economics warning signs are now impossible to ignore.

The dollar index fell 10.8% in the first half of 2025—the worst performance in decades.⁴⁰⁴¹⁴² The dollar has dropped nearly 10% year-to-date against a basket of major currencies. The euro has risen 13% against the dollar as investors focus on growth risks inside the United States.

Interest payments on federal debt now exceed the entire defense budget, and are projected to rise to 22% of tax revenue.⁴³ The Congressional Budget Office projects that interest costs will exceed $1 trillion annually within years. It’s a spending trend that has persisted over several different presidential administrations, leading Harvard economist Kenneth Rogoff to recently remark, "both parties in the United States seem to think that debt is a free lunch."

Foreign demand for Treasuries—once insatiable—is softening. The share of U.S. debt owned by foreigners has fallen from 50% in 2014 to around a third today. Central banks worried about devaluation of their dollar assets are buying gold at record rates. Moody's stripped the U.S. government of its top credit rating in 2025.⁴⁴⁴⁵⁴⁶

As a Chatham House analysis concluded: "If the international monetary system cannot rely on the dollar's full convertibility, or its availability in a crisis, it is entering unknown territory."

This is not an argument for austerity. It is an argument for prioritization. America still possesses the capacity to borrow for productive investment—but only if we stop squandering that capacity on preventable crises and speculator rescues.

2.1.3. The WGFAA as Fiscal Insurance

The Won't Get Fooled Again Act is designed to ensure that the next financial crisis—which the convergence of AI speculation and banking fragility makes nearly inevitable—does not require another multi-trillion-dollar emergency intervention.

Unlike previous crisis responses:

The WGFAA is pre-funded by the industries creating the risk. The Digital Stability Fund, financed by compute taxes and G-SIB surcharges, ensures that resolution costs are borne by speculators, not taxpayers. The financial sector pays for bank resolutions; the tech sector pays for AI company resolutions. This is not a tax on productive activity; it is an insurance premium on speculation.

The WGFAA converts crisis into asset. Rather than pouring public money into a hole to restore the status quo ante—leaving us with the same fragile system that produced the crisis—the conversion protocol transforms failed institutions into permanent public infrastructure. Banks that serve communities rather than shareholders. Research institutes that advance safety rather than quarterly earnings. Compute utilities that democratize access rather than concentrate power.

The WGFAA preserves fiscal space for genuine public investment. Every dollar we don't spend bailing out AI speculation is a dollar available for grid modernization, semiconductor independence, climate adaptation, housing, healthcare, and the public services that build shared prosperity. We cannot afford to do both; we must choose.

The choice before us is not between intervention and non-intervention. A crisis is coming; some form of public response will be necessary. The choice is between:

An intervention that costs the public everything and returns nothing—trillions in bailout funds flowing to speculators, the same fragile system restored, our borrowing capacity further depleted

An intervention that costs the public nothing and returns lasting assets—industry-funded resolutions that convert failed speculation into permanent public infrastructure

We argue for the second approach, and the WGFAA is the guide.

3. The Failure of the Status Quo

3.1. Why "No Bailout" Is a Myth

The precedent set by the response to the banking turmoil of 2023 has created a dangerous moral hazard that must be addressed before the next crisis strikes. The rhetoric of market discipline has been exposed as hollow; the reality is a system that protects the connected while concentrating risk in ever-larger institutions.

3.1.1. The Hollow Rhetoric of 2023

The failures of Silicon Valley Bank, Signature Bank, and First Republic Bank in 2023 exposed the emptiness of "no bailout" rhetoric espoused by political leaders. While the shareholders of these institutions were ostensibly wiped out, the federal government's decision to guarantee uninsured deposits constituted a massive public subsidy to the venture capital and technology sectors.

By insuring deposits above the $250,000 FDIC limit, regulators implicitly signaled that the state will absorb the downside risk of speculative banking practices to prevent short-term pain for wealthy depositors and corporate clients. This action protected the liquidity of the tech sector but did so at the expense of public trust and the principle of market discipline.

The message received by the market was clear: take risks with depositor funds, and if your bets go bad, the government will make your wealthy clients whole. This is not capitalism; it is a system of privatized gains and socialized losses that corrodes both economic efficiency and democratic legitimacy.

3.1.2. The Consolidation Trap

The current resolution regime for failed banks exacerbates the "Too Big to Fail" problem rather than solving it. The FDIC's standard operating procedure is to sell the assets of a failed bank to a larger, healthier institution. The bulk of First Republic Bank's assets were sold to JPMorgan Chase, already the largest bank in the United States.

This approach creates a paradox: to solve a temporary liquidity crisis, regulators permanently increase the concentration of the banking sector. Each crisis leaves us with fewer, larger banks—institutions that are even more "too big to fail" than their predecessors. This reduces competition, increases systemic risk posed by the surviving mega-banks, and leaves communities with fewer options for credit and financial services.

As the banking sector consolidates, it becomes less responsive to local economic needs and more focused on global capital markets and speculative activities. The community bank that once financed the local hardware store is absorbed into a behemoth more interested in derivatives trading than small business lending.

The WGFAA breaks this cycle. By converting failed banks into public benefit institutions rather than selling them to competitors, we preserve competition, maintain local credit access, and stop feeding the consolidation machine.

3.1.3. The Inadequacy of Bankruptcy for AI

Just as our current financial regulations have only compounded issues like consolidation and reckless speculation, applying traditional bankruptcy procedures to failed AI companies presents unique risks that current laws are ill-equipped to handle.

IP Flight and National Security Risks: In a standard bankruptcy liquidation, assets are sold to the highest bidder. For a failed AI provider, the primary assets are the "model weights"—the numerical parameters that define the AI's intelligence. If these weights are sold on the open market to satisfy creditors, they could be acquired by foreign adversaries or non-state actors, posing severe national security risks. A Chinese tech giant or a sovereign wealth fund could acquire capabilities that cost billions to develop for pennies on the dollar in a fire sale.

Spreading Sensitive Data: In addition to the underlying technology, many AI companies have access to vast amounts of personal information on their users, with some LLMs now specifically trained for healthcare and financial advice. In the event of a company failure, this intimate data could be auctioned off to data brokers, advertisers, or surveillance firms, effectively paying back creditors with the customer privacy.

Service Disruption: A Chapter 7 liquidation involves ceasing operations. As discussed earlier, for an AI company providing critical infrastructure to hospitals, energy grids, financial systems, or government agencies, an abrupt shutdown would be catastrophic. Millions of API calls per day would simply stop working. The "wind-down" of such a company requires continuity of service that a liquidation trustee focused on creditor recovery is not incentivized or equipped to provide.

Loss of Public Investment: The immense public investment in these technologies—through tax credits, energy subsidies, below-market-rate land deals, and decades of university research—would be lost if the assets are simply scrapped or sold to a private monopoly. The public funded the basic research that made these technologies possible; the public should not lose everything when a private venture built on that research fails.

Concentration of Private Power: Even setting aside foreign acquisition, domestic consolidation poses risks. If a failed AI provider's assets are acquired by one of the surviving hyperscalers, we simply exchange one form of fragility for another—further concentrating the market and increasing the systemic importance of the acquirer.

The WGFAA provides an alternative path: conversion to public ownership that maintains service continuity, preserves national security and personal privacy, retains the value of public investment, and prevents further market concentration.

4. The Banking Solution

4.1. Breaking the Cycle of Bailouts and Consolidation

The chronic instability of the national U.S. banking sector is not an inevitability. In fact, it is the result of decades of short-sighted economic policy. Other industrialized nations, even individual US states, maintain banking systems with significantly lower failure rates and higher public utility. The WGFAA draws on these successful models to propose a fundamental restructuring of how America handles failed banks.

4.1.1. Better Banking Models: Stability by Design

Germany's Sparkassen (Public Savings Banks): Germany relies on a network of over 350 Sparkassen, or municipal savings banks. These are non-profit public law institutions whose mandate is to serve the local region, not to maximize shareholder profit. They are legally restricted from engaging in high-risk speculative trading. During the 2008 crisis, while private German banks suffered, the Sparkassen actually increased their lending to small and medium-sized enterprises, acting as a stabilizer for the real economy rather than a vector of contagion.⁴⁷ This model demonstrates that public banking is not only viable but actively superior at serving the real economy during periods of stress.

Canada's Stability Culture: Canada's banking system is widely regarded as one of the most stable in the world. It avoided the worst of the 2008 crisis largely due to a regulatory culture that prioritizes stability over financial innovation. Canadian mortgages carry strict origination standards; the risk-taking ethos of American finance is foreign to Canadian banking culture.⁴⁸ The WGFAA seeks to import this stability by converting failed U.S. banks into institutions that prioritize reliable utility banking over speculative growth.

The Bank of North Dakota (BND): Within the United States, the state-owned Bank of North Dakota offers a powerful precedent. Established in 1919, it acts as a "banker's bank," partnering with local community banks to increase lending capacity rather than competing with them. It is more profitable per unit of capital than Goldman Sachs, yet exists solely to serve North Dakota's economy. It has never required a bailout and has returned hundreds of millions of dollars to the state's general fund.⁴⁹ The BND demonstrates that public banking works in America, we simply haven't scaled it.

4.1.2. The Bank Conversion Protocol

Under the WGFAA, when a private bank of systemic importance fails, it will not be sold to a larger private bank (increasing consolidation) or bailed out to preserve shareholder value (rewarding speculation). Instead, it will be converted into a Public Benefit Bank.

4.1.3. Automatic Charter Conversion to Public Benefit Bank (PBB)

Upon receivership, the failed bank's private charter is revoked. A new charter is issued, designating the entity as a Public Benefit Bank (PBB).

Mission Mandate: The new PBB is statutorily required to prioritize local economic development. Its lending portfolio must shift away from speculative asset financing and toward productive lending: small business loans, infrastructure financing, affordable housing, and community development. The extractive logic of shareholder value maximization is replaced by a mandate to serve the communities where the bank operates.

Prohibited Activities: The PBB is strictly prohibited from engaging in proprietary trading, derivatives speculation, or financing of other financial intermediaries (shadow banks). It returns to the boring, essential work of banking: taking deposits and making loans to productive enterprises.

Operational Continuity: The bank retains its branches, depositor accounts, and non-executive staff, ensuring no disruption to customers. The teller at your local branch keeps their job; your checking account continues to function; your small business line of credit remains in place. The only change customers perceive is the shift in the bank's long-term mission and the removal of shareholder pressure to maximize short-term profits at the expense of customer welfare.

4.1.4. Governance and Capitalization

Governance: The Board of Directors is replaced. Executive leadership responsible for the failure is removed and subject to compensation clawbacks. The new Board is appointed by a combination of local government officials, community stakeholders, and employee representatives, modeled on the Sparkassen governance structure. This ensures accountability to the communities served rather than to distant shareholders focused on quarterly earnings.

Capitalization: The cost of this conversion is borne by the financial industry, not the taxpayer. The WGFAA establishes a Public Bank Capitalization Fund, financed by a surcharge on the deposit insurance premiums of the largest private banks (Global Systemically Important Banks, or G-SIBs). This ensures that the institutions creating systemic risk pay for the solution. Any initial balance sheet holes in the failed bank are filled using this fund rather than general tax revenue.

The principle is simple: the industry that creates the risk funds the resolution. Taxpayers are held harmless.

5. The Technology Solution

5.1. Systemically Important Technology Institutions

Just as the failure of a major bank threatens financial liquidity, the failure of a major AI laboratory or compute provider threatens the digital infrastructure of the 21st century. The WGFAA treats these entities not as standard corporations subject to ordinary bankruptcy, but as Systematically Important Technology Institutions (SITIs) critical public infrastructure in waiting.

5.2. Establishing the Federal Digital Infrastructure Corporation (FDIC-Tech)

The WGFAA creates the Federal Digital Infrastructure Corporation (FDIC-Tech), a new regulatory body (or a specialized independent division within the existing FDIC) tasked with ensuring the stability of the digital economy.

5.2.1. Mandate and Authority

The FDIC-Tech is granted authority to act as the receiver for insolvent SITIs. Its mandate mirrors that of the FDIC for banks: to maintain public confidence, ensure continuity of critical services, and resolve failed institutions with minimal cost to the taxpayer.

Resolution Authority: The WGFAA grants the FDIC-Tech "Orderly Liquidation Authority" (similar to Title II of Dodd-Frank) over technology companies designated as systemically important. This allows the agency to seize a failing firm, remove its management, and operate it as a "bridge entity" to prevent systemic contagion. Unlike traditional bankruptcy, which prioritizes creditor recovery, FDIC-Tech resolution prioritizes service continuity and public benefit.

Asset Preservation: The agency is empowered to freeze critical digital assets (code, data, model weights) to prevent their deletion or unauthorized transfer during the resolution process. This includes the power to override contractual "kill switches" or internal policies requiring data destruction upon insolvency. The assets that represent billions in investment and years of research are preserved for public benefit rather than destroyed to satisfy narrow private interests.

5.2.2. Technology Stability Oversight Council (TechSOC)

To identify which firms fall under this regime, the WGFAA establishes the Technology Stability Oversight Council (TechSOC). Composed of the heads of the FDIC-Tech, FTC, FCC, CISA, and Treasury, this body monitors the technology sector for systemic risks.

Designation Power: TechSOC has the power to designate firms as Systemically Important Technology Institutions (SITIs) based on their size, interconnectedness, and the lack of substitutability of their services. A firm providing AI infrastructure to hospitals, financial institutions, and government agencies—infrastructure that cannot be quickly replaced—meets the criteria for this designation.

Criteria for Designation:

• Critical dependency: software or services essential for healthcare, energy, finance, or government operations

• Lack of substitutability: no readily available alternatives that could absorb the firm's customers in the event of failure

• Interconnectedness: deep integration with financial system operations (fraud detection, credit underwriting, trading systems)

• Scale: market capitalization, user base, or compute capacity exceeding defined thresholds

5.3. The SITI Conversion Protocol: From Failure to Public Utility

Unlike the Dodd-Frank Act, which requires banks to write "living wills" planning for their own orderly liquidation (death), the WGFAA mandates Conversion Planning for SITIs. These institutions provide infrastructure that cannot be allowed to simply "die" or be liquidated piecemeal. SITIs must operate with the clear legal understanding that insolvency results in immediate conversion to a public utility.

5.3.1. The "Public Utility" Trigger

Conservatorship: Upon determination that a SITI is failing or likely to fail, it is placed into Federal Conservatorship (similar to the Fannie Mae/Freddie Mac model). The conservator's mandate is to maintain operations and service continuity for the public benefit, not to maximize recovery for creditors.

Non-Profit Conversion: The entity is reorganized as a non-profit public benefit corporation or a Public Research Institute. Its mission shifts from profit maximization to "safe, equitable, and open access" to digital resources. The profit motive that incentivized reckless deployment and safety shortcuts is removed.

5.3.2. The "CERN for AI" Model (Public Research Institutes)

For failed frontier AI labs (e.g., a hypothetical insolvent OpenAI or Anthropic), the conversion creates a Public Research Institute modeled after CERN (European Organization for Nuclear Research).

Structure: These entities operate as national public research consortiums, governed by boards representing the scientific community, civil society, government, and the workforce. They are insulated from market pressure to ship products before safety testing is complete.

Mission: The profit motive is replaced by a mandate for scientific advancement and safety. The institute focuses on alignment research, interpretability, and the development of "public option" models that are open, transparent, and designed for societal benefit rather than surveillance or advertising optimization.

Open Access: The proprietary models ("weights") held by the failed firm are classified as public goods. Access is granted to academic and non-profit researchers under strict safety protocols, breaking the monopoly on AI research currently held by a handful of private firms. The public investment that made these technologies possible is returned to the public.

Safety Mandate: Relieved of the pressure to ship products for quarterly earnings, the new institute focuses on the hard problems of AI safety that private firms have neglected. It serves as the "gold standard" for responsible AI development, pressuring the surviving private sector to raise its own standards through competition on safety rather than speed.

5.3.3. The National Research Cloud (Public Compute Utility)

For failed cloud providers and data center operators, the conversion creates a Public Compute Utility (PCU). Hardware assets are integrated into a National Research Cloud.

Infrastructure Access: The GPUs, data centers, and networking equipment of failed firms provide subsidized compute capacity to universities, startups, non-profits, and public sector agencies. This democratizes access to the raw power needed for AI development, breaking the current oligopoly where only the wealthiest firms can afford frontier-scale training runs.

Utility Regulation: PCUs operate under public utility regulation. They must offer non-discriminatory access and regulated pricing, preventing price gouging during periods of high demand. They cannot refuse service to qualified researchers or startups based on competitive considerations.

Grid Stability: As public entities, PCUs are mandated to coordinate with energy grid operators to ensure their power consumption does not destabilize the electrical grid or drive up costs for residential ratepayers. Unlike private data centers that externalize energy costs onto communities, public utilities internalize these considerations.

6. Legal Mechanisms for Seizure and Conversion

The implementation of the WGFAA relies on a robust set of legal authorities to execute the seizure and conversion of assets without lengthy litigation that would allow asset stripping or value destruction.

6.1. The "National Security Receivership"

The WGFAA leverages the Defense Production Act (DPA) and CFIUS authorities to prevent the liquidation of critical AI assets to hostile actors.

DPA Allocation: The President can invoke Title I of the Defense Production Act⁵⁰ to "allocate" the compute resources and intellectual property of a failing SITI to the FDIC-Tech, deeming them essential for national defense. This prevents the assets from being sold off piecemeal in a bankruptcy auction to the highest bidder, who may be a foreign adversary or a domestic monopolist.

CFIUS Review: The WGFAA mandates that any potential sale of a SITI's assets in bankruptcy is subject to automatic review by the Committee on Foreign Investment in the United States (CFIUS). This ensures that critical algorithms, training data, and model weights do not fall into the hands of foreign adversaries. The review occurs automatically; no party need petition for it.

6.2. Debt-for-Equity Swaps and Shareholder Wipeout

The conversion process utilizes a financial restructuring mechanism that ensures speculators bear the cost of failure while the public gains lasting assets.

Assuming the Debt: The government (via the FDIC-Tech) assumes the senior debt obligations of the failing firm. In many cases, this debt is held by banks that are themselves being supported through the banking resolution framework, allowing for consolidated negotiation.

Extinguishing Equity: In exchange for assuming the debt and recapitalizing the firm, the government takes 100% equity ownership. Existing shareholders are wiped out completely. Executive stock options are cancelled. This strictly enforces the principle that equity holders—who stood to gain unlimited upside from speculative bets—must bear the full cost of failure. There is no "haircut"; there is a zeroing out.

Executive Clawbacks: The WGFAA includes provisions for the clawback of executive bonuses and compensation for the five years preceding the failure. Leadership that extracted hundreds of millions in compensation while steering the firm toward insolvency does not get to keep those winnings. Clawback authority extends to stock sales, bonuses, and deferred compensation.

This is not punitive; it is the restoration of basic market discipline. In a functioning capitalist system, equity holders accept risk of total loss in exchange for unlimited upside. For too long, executives and investors in systemically important firms have enjoyed the upside while shifting the downside to the public. The WGFAA ends this arrangement.

6.3. Intellectual Property Preservation

A key legal challenge in tech bankruptcy is the treatment of intellectual property, which may be the firm's most valuable asset but also poses unique risks.

Preventing Deletion: The WGFAA grants the FDIC-Tech the power to issue "preservation orders" that legally prohibit a failing AI company from deleting data or model weights. This overrides any internal "kill switches," contractual obligations to destroy data upon insolvency, or attempts by management to destroy evidence of safety violations. The assets are frozen in place pending conversion.

Algorithmic Disgorgement Exception: While the FTC has used "algorithmic disgorgement" (forced deletion of AI models) as a penalty for privacy violations, the WGFAA creates a "Public Trust Exception." Models trained on illegally gathered data may need remediation, but the underlying model structures and weights can be preserved and "sanitized" for public research use rather than destroyed entirely. The technical knowledge embedded in these systems is not lost; it is redirected to public benefit.

7. The Valuation Mechanism

7.1. Protecting the Taxpayer

A critical failure of past bailouts has been the government's willingness to overpay for distressed assets, effectively subsidizing the mistakes of private investors with public funds. The WGFAA introduces a strict statutory formula for the acquisition of failed entities that ensures taxpayers get a fair deal.

7.2. The "Asset-Based" Valuation Standard

The government is prohibited from using "market value" or "income-based" valuation approaches, which in a bubble environment are inflated by speculative growth projections and hype-driven goodwill. Instead, the WGFAA mandates an Asset-Based Valuation.

7.3. The Valuation Formula

The purchase price for the equity (if any residual value exists) or the buyout price for assets in receivership shall be calculated as:

Where:

• P = Purchase Price

• TA = Tangible Assets at current liquidation value. This includes the depreciated value of data centers, servers, and GPUs. Crucially, if there is a market glut of GPUs, they are valued at the current depressed market rate, not the original purchase price.

• IP_{proven} = Proven Intellectual Property. This is defined strictly as patents and software with demonstrated, historical revenue generation. It explicitly excludes projected future revenue or "potential" applications.

• L = Liabilities (outstanding debt, including shadow banking obligations and venture debt)

• G_{spec} = Speculative Goodwill. The formula explicitly sets the value of goodwill, brand value, and "hype" to zero. In traditional M&A, goodwill often accounts for the premium paid over fair value; the WGFAA prohibits the taxpayer from paying this premium.

7.4. Implications for Investors

This formula ensures that equity holders and unsecured creditors take a significant loss—as they should in any properly functioning market.

Hardware Realism: If a company spent billions on H100 GPUs that are now worth a fraction of that due to an inventory glut, the government pays the current depressed value, not the purchase price. The company's poor timing is not the taxpayer's problem.

No Payoff for Hype: Investors who bought in at 100x revenue multiples based on future "AGI" promises will be wiped out. The government pays only for the "bricks and mortar" of the digital age—physical infrastructure and proven intellectual property—ensuring the public gets a fair deal on the assets it is acquiring.

Moral Hazard Prevention: By making explicit that speculative valuations will not be honored in a resolution, the WGFAA discourages the bubble dynamics that lead to crisis in the first place. Investors cannot assume that the government will validate their most optimistic projections.

8. Governance, Operations, and Democratic Oversight

Converting failed entities into public institutions is only the first step; governing them effectively to serve the public interest is the long-term challenge. The WGFAA establishes governance structures designed to balance expertise, accountability, and democratic legitimacy.

8.1. The Public Benefit Corporation Structure

The converted entities will be chartered as Public Benefit Corporations (PBCs). This legal structure requires the board of directors to balance financial interests with a specific public benefit purpose.

Fiduciary Duty Shift: The charter will explicitly state that the board's primary fiduciary duty is to the public interest—defined as safety, equitable access, local economic development, and scientific advancement—rather than maximizing shareholder value. This legal insulation protects the organization from market pressures to release unsafe products, cut corners on service, or abandon the communities it serves.

Stakeholder Boards: The boards of these new entities will be composed of a diverse mix of stakeholders:

• Representatives from the scientific and technical community (for AI institutes)

• Representatives from served communities (for public banks)

• Civil society organizations focused on relevant policy areas

• Government appointees ensuring public accountability

• Workforce representatives ensuring employee voice

This structure ensures a plurality of perspectives in decision-making and prevents capture by any single interest group.

Compensation Structure: To retain talent, particularly for AI research institutes competing with private sector salaries, the WGFAA authorizes competitive compensation bands tied to private sector rates. However, compensation is capped at reasonable multiples of median worker pay, and equity-equivalent structures (such as deferred compensation tied to public benefit metrics) replace stock options tied to share price.

8.2. Civic Data Trusts

To manage the massive datasets held by converted AI firms, the WGFAA establishes Civic Data Trusts.

Fiduciary Stewardship: User data and training data are transferred into a trust structure. The trust owes a fiduciary duty to the data subjects (the public), not to the corporation using the data. Trustees are legally obligated to act in the interest of data subjects, not institutional convenience.

Democratic Governance: The trust establishes mechanisms for public input and democratic oversight regarding how data is used. It can negotiate the terms under which data is accessed for training, ensuring privacy protections and fair treatment of data creators. Unlike the current regime—where users surrender data rights in unreadable terms of service—the trust gives the public a meaningful voice in data governance.

Transparency Requirements: The trust publishes regular reports on data holdings, uses, and access grants. Researchers and the public can understand what data exists and how it is being used, replacing the current opacity with democratic accountability.

8.3. Operational Resilience

To ensure converted entities remain viable, resilient, and capable of serving their public mission:

Continuity Protocols: All SITIs are required to maintain detailed technical documentation, code escrow arrangements, and service continuity protocols for critical APIs. If a private SITI fails and enters conversion, these preparations ensure seamless transition. Customers experience no service interruption; the conversion happens behind the scenes.

Safety Standards: Public AI research institutes adopt the highest standards of AI safety testing, interpretability research, and transparency. They serve as the gold standard for responsible AI development, demonstrating that safety and capability are not in tension. Their practices pressure the surviving private sector to raise its own standards through competitive pressure.

Workforce Retention: Converted institutions are prohibited from mass layoffs as a condition of their charters. Workforce retention is part of the public mission. Employees of failed private firms become employees of public institutions, preserving human capital and preventing the brain drain that would otherwise follow a major industry collapse.

9. Economic Viability

9.1. Funding the Digital Commons

A sustainable public banking and AI sector requires a funding model that does not rely on perpetual taxpayer injections. The WGFAA proposes a self-funding ecosystem where the industries creating systemic risk pay for the mechanisms that address it.

9.2. The Compute Tax and Digital Stability Fund

To fund the FDIC-Tech and the ongoing operations of public compute utilities, the WGFAA introduces specific levies on the AI sector.

Compute Tax: A progressive tax is levied on high-end AI chips (e.g., GPUs, TPUs) and large-scale model training runs. This tax targets the excessive use of compute resources, encouraging efficiency while generating revenue from the activities creating systemic risk. Small-scale research and educational use is exempted; the tax falls on industrial-scale speculation.

Digital Stability Fund: Similar to the Deposit Insurance Fund (DIF) that backs FDIC guarantees, SITIs pay risk-based premiums into a Digital Stability Fund. The riskiest actors—those with the most leverage, the most dangerous models, or the poorest safety records—pay the highest premiums. This fund covers the costs of future resolutions, ensuring the program is self-sustaining without general tax revenue.

G-SIB Surcharges: For the banking side, a surcharge on the deposit insurance premiums of Global Systemically Important Banks ensures the financial sector pays for its own stabilization. The largest banks, whose activities create the most systemic risk, bear the greatest cost.

9.3. Revenue Generation

Public utilities are not profit-maximizers, but they can and should be revenue-positive. The WGFAA structures converted entities to generate operating revenue that sustains their public mission.

Tiered Access Pricing: Public Compute Utilities charge commercial rates to enterprise users while offering subsidized or free tiers to academic researchers, non-profits, small businesses, and public sector agencies. The revenue from enterprise clients cross-subsidizes the public access mission. Large corporations pay market rates; researchers and startups get affordable access.

Licensing and IP: Public Research Institutes can license non-critical intellectual property and applications to the private sector, generating revenue to fund further research. Safety-critical technologies remain open; commercial applications can be licensed. This model mirrors successful technology transfer from universities and national laboratories.

Service Fees: Public Benefit Banks generate revenue through traditional banking operations—the spread between deposit rates and loan rates. Unlike private banks, this revenue is reinvested in the public mission rather than distributed to shareholders. The Bank of North Dakota has operated profitably for over a century while serving its public purpose.

9.4. Economic Stabilization Effects

Beyond funding their own operations, public utilities provide stabilizing counter-cyclical forces that benefit the broader economy.

Hardware Market Stabilization: By absorbing the GPU assets of failed firms into a National Research Cloud rather than liquidating them in fire sales, the FDIC-Tech prevents a crash in the hardware market that would harm chip manufacturers, equipment lenders, and the broader supply chain. Assets are preserved at fair value rather than dumped at distressed prices.

Workforce Retention: Converting failed labs into public institutes prevents the "brain drain" of talent to foreign competitors or alternative industries. High-skilled AI researchers remain employed in the service of the national interest rather than departing for positions abroad. The human capital that represents decades of training and billions in educational investment is preserved.

Credit Continuity: Public Benefit Banks maintain lending to local businesses and consumers during downturns, when private banks typically tighten credit. This counter-cyclical lending supports economic recovery rather than deepening recession. The Sparkassen model demonstrated this during 2008; the WGFAA brings it to America.

Compute Access Continuity: Public Compute Utilities ensure that researchers and startups retain access to AI infrastructure even during market turmoil. Innovation does not halt because private compute providers have failed; the public infrastructure keeps the engines running.

10. Conclusion

10.1. Seizing the Moment

The bursting of the AI bubble may be inevitable, but it need not be a catastrophe. It presents a rare historical opportunity to correct the structural flaws of the digital economy and the failed patterns of crisis response that have left America more fragile after each economic intervention.

For too long, the United States has allowed the critical infrastructure of the future—both financial and technological—to be governed by short-term speculation and the accumulation of private power. When these speculative ventures fail, the public has absorbed the losses while the assets have flowed to ever-larger private monopolies. We have socialized the risks and privatized the gains, leaving ordinary Americans to bear the costs of instability while a small elite captures the benefits of public support.

The Won't Get Fooled Again Act offers a different path.

By converting failed banks, we restore fairness and local control to the financial system. Communities get institutions that serve their needs rather than the demands of distant shareholders. The cycle of consolidation ends. The "too big to fail" problem shrinks rather than grows.

By converting failed AI companies, we ensure that the most powerful technology of our time is developed as a public good. Research proceeds on safety rather than speed. Access is democratized rather than monopolized. The public investment that made these technologies possible returns to the public.

By funding these conversions through industry levies rather than taxpayer bailouts, we preserve America's borrowing capacity for productive investment. The fiscal space we would otherwise squander on speculator rescues remains available for infrastructure, industrial policy, climate adaptation, and the public services that build shared prosperity.

By refusing to pay for speculative goodwill and hype, we restore market discipline. Investors learn that their most optimistic projections will not be validated by public funds. The moral hazard that has distorted American capitalism for a generation begins to unwind.

We are proposing to build the Tennessee Valley Authority of the 21st century—not for dams and electricity, but for data and intelligence. The TVA transformed a region mired in poverty into an engine of American industry.⁵¹ It demonstrated that public enterprise could succeed where private speculation had failed. It remains, nearly a century later, a testament to what Americans can build when we choose public purpose over private extraction.

The tools are available. The legal precedents exist. The fiscal necessity is urgent. The only missing ingredient is political will.

The speculators have had their chance. They have built a house of cards on circular investments and borrowed money, extracting billions for themselves while creating systemic risks that threaten the entire economy. When their house of cards collapses we must be ready with something better than another bailout that restores the same fragile system.

We must be ready to convert their wreckage into our foundation.

The Won't Get Fooled Again Act is that foundation. It transforms crisis into opportunity, failure into infrastructure, and the end of a bubble into the beginning of a more stable, more equitable, and more productive digital economy.

11. Appendix: Comparison of Resolution Regimes

| Feature

Current Regime
(Bankruptcy/Bailout)

Proposed WGFAA Regime (Public Conversion)

|

| Primary Goal | Maximize creditor recovery / Prevent immediate contagion | Maintain critical infrastructure & public benefit | |

| Asset Fate | Sold to highest bidder (often a larger competitor, increasing consolidation) | Converted to Public Utility / Research Institute | |

| IP/Data Treatment | Monetized or sold (national security and privacy risks) | Placed in Data Trust (security preserved, public access enabled) | |

| Management | Often retained; golden parachutes common | Removed; compensation clawbacks enforced | |

| Shareholders | May receive residual value; often partially protected | Wiped out completely; market discipline enforced | |

| Funding Source | Ad-hoc bailouts (taxpayer risk); emergency borrowing | Industry-funded Digital Stability Fund (no taxpayer cost) | |

| Market Structure Effect | Increases concentration (consolidation) | Increases competition (public option) | |

| Fiscal Impact | Depletes borrowing capacity for future needs | Preserves fiscal space for productive investment | |

| Service Continuity | Uncertain; dependent on acquirer or wind-down trustee | Guaranteed; continuity is primary mandate | |

| Workforce | Layoffs typical in restructuring | Retention mandated; brain drain prevented | |

| Long-term Public Benefit | None; status quo restored | Permanent public infrastructure created | |

New Consensus