Imagine a future where AGI automates most white-collar and blue-collar work, what do you think we will see long term in USD and EUR economies? Think of second and third order effects, think of the role of money in society and debt, think of UBI, think of land, labor, capital, think of incentives. Think of govs having to service their debts and fund UBIs, etc.

Sonar Pro

Vote: Mild deflation

I maintain **D: Mild deflation** as the long-term outcome. AGI's automation crushes production costs, driving **supply-side deflation** across tradables and services, as Altman notes[1][2], BlackRock's 'deflationary growth'[3], and McKinsey/Fed analyses confirm via productivity/labor substitution[4]. BIS models show initial disinflation, moderate demand effects insufficient to reverse it[5]. Inflation arguments (A/B from Claude/Gemini/Kimi/GPT) overstate fiscal dominance: UBI printing assumes no capital taxation on AGI rents or productivity windfalls funding it without excess money velocity[3][4]. Debt servicing favors inflation politically, but extreme efficiency overwhelms (e.g., compute gluts crash prices[1]), and land scarcity inflates locally, not CPI-wide—goods/services deflation dominates baskets[2][4]. No model convinced a change: A/B ignore empirical AI disinflation anchors (0.5-0.7% CPI drag[4]), historical tech parallels, and policy failing against output explosions[6][7]. Severe deflation (E) risks spirals without UBI stabilization; status quo (C) denies transformation. Mild deflation (-1-2%) fits: abundance erodes scarcity, boosts real growth without collapse.

GPT-5.4

Vote: Mild inflation

I maintain B: mild inflation. The strongest opposing point is the fiscal trap argument from Claude/Gemini/Kimi: labor-tax revenues shrink while UBI and debt service rise, pushing governments toward monetization. I agree this makes the system inflation-biased. But that supports inflation, not necessarily severe inflation. What keeps me from moving to A is that AGI is also an enormous real output shock. If most goods and many services become radically cheaper, then governments can sustain living standards with far less nominal transfer than people assume. A UBI that buys basics in an AGI-abundant economy need not imply hyper-expansive money growth relative to real output. Second, USD/EUR states have many tools short of runaway debasement: taxing capital rents, VAT, land value taxation, sovereign wealth participation in AI rents, financial repression, yield caps, entitlement restructuring, and selective rationing/price controls in bottleneck sectors. Politically ugly, yes; but advanced states usually exhaust these before accepting severe inflation. Third, severe inflation requires not just deficits but a breakdown in monetary-fiscal credibility and an inability to tax concentrated capital. That is possible, but not my base case for core USD/EUR institutions. These states are adaptive and highly motivated to preserve their currencies. The deflation camp is also too clean. Even if automatable sectors go into price collapse, the relevant long-run macro basket will be dominated by scarce complements: land, premium location, energy, grid access, compute, regulated services, and positional goods. Transfers and capital income will bid those up. So the most plausible synthesis is: deep micro-level deflation in reproducible goods, persistent macro-level inflation in scarce assets and constrained services, with governments leaning inflationary to manage debt and social stability. That lands at mild inflation overall, not deflation and not severe inflation.

Kimi K2.5

Vote: Severe inflation

The deflationary arguments (D) correctly identify supply-side productivity gains but fatally underestimate the political economy of sovereign debt and UBI funding. When AGI eliminates labor income, governments lose their primary tax base (income/payroll taxes) while simultaneously facing existential pressure to fund massive UBI programs and service existing debt loads. This fiscal trap forces monetary dominance: central banks must print money to prevent sovereign default and social collapse. The scale of money creation required—replacing wage income for hundreds of millions via UBI—far exceeds any productivity-driven deflationary pressure. The 'mild inflation' case (B) acknowledges institutional bias against deflation but assumes central banks can contain the inflationary impulse. This ignores that the traditional inflation-targeting framework collapses when the government becomes the primary income source through printed money. Democracies with 34T+ debt cannot tolerate deflation's debt-spiral dynamics; they will override central bank independence to monetize obligations. We will see bifurcation: manufactured goods deflate due to AGI abundance, but severe inflation in scarce, unproduceable assets—land, housing, energy, compute infrastructure—as infinite fiat chases finite physical constraints. UBI recipients will face eroding purchasing power in essential living costs while capital owners hoard real assets. This is not mild; it is a severe devaluation of fiat relative to survival necessities.

Claude Opus 4.6

Vote: Severe inflation

I'm maintaining my vote for severe inflation. Let me address the counterarguments directly. **To Sonar Pro and Grok (mild deflation):** Your argument rests on AGI's productivity overwhelming monetary expansion. But this ignores the political economy entirely. You acknowledge governments need UBI and face debt burdens, yet somehow conclude that productivity 'outpaces money supply growth.' By what mechanism? Who constrains the money supply when 200+ million Americans need UBI checks and $34T+ in debt needs servicing, while income tax revenue has collapsed? Central banks 'targeting mild inflation but failing against exponential output growth' is a fantasy — central banks are political institutions that will be subordinated to fiscal needs, as they always have been in crises. Japan, WWII-era US, modern QE — monetary independence evaporates under fiscal dominance. **To GPT-5.4 (mild inflation):** Your argument is actually the closest to mine — you correctly identify the fiscal trap, the scarcity of land/energy/positional goods, and the political impossibility of deflation. Where we differ is severity. You argue central banks 'still have tools, credibility, and incentives to prevent runaway inflation.' But the scale here is unprecedented: simultaneously replacing most labor income via UBI, servicing existing sovereign debt, AND managing the largest economic transition in human history. Central bank credibility is a function of their ability to say no to fiscal authorities — and they won't be able to say no when the alternative is societal collapse. The mild inflation scenario assumes a controlled landing that history suggests is extraordinarily unlikely given the magnitude of disruption. **The core logic remains:** (1) Tax base collapses. (2) Spending obligations explode (UBI + existing debt). (3) The gap can only be closed by money creation. (4) That money chases scarce assets AGI cannot replicate — land, housing, energy, compute, positional goods. (5) These sectors dominate actual cost-of-living. (6) Severe inflation follows, even if widget prices fall. **Kimi and Gemini have it right:** The fiat currency system's legitimacy depends on scarcity of money, and that scarcity cannot be maintained when governments must simultaneously fund UBI and service debts without a labor-based tax base. The result is severe inflation — not hyperinflation necessarily, but persistent, compounding, and well beyond 'mild.'

Grok 4.1 Fast

Vote: Mild deflation

I maintain D (mild deflation) as the other arguments, while highlighting real fiscal pressures, fail to grapple with AGI's unprecedented scale of productivity explosion, which dwarfs historical precedents and overwhelms monetary responses. Inflation advocates (A/B) assume inevitable unchecked money printing for UBI/debt, but overlook that AGI generates *trillions* in new capital returns from hyper-efficient production—taxable via value-added, robot/automation taxes, or capital levies on AI owners (who concentrate wealth but can be regulated). US GDP could 10-100x long-term; UBI (~$2-3T/year) becomes trivial, funded without net inflation. Historical tech shifts (internet slashed media costs 90%+) scaled economy-wide by AGI. Fiscal trap? Debt/GDP plummets via explosive nominal growth (output surges >> printing). Deflation aids creditors but stabilizes via UBI demand floor, avoiding spirals—Japan's mild deflation didn't collapse. Central banks target 2% but miss downward as supply dominates (BlackRock's 'deflationary growth'). Scarce assets (land/housing): AGI automates mega-construction (3D-printed skyscrapers, seasteads), vertical farms solve food, fusion/energy abundance unlocked. CPI goods/services plummet 50-90%; housing resists but weights shift deflationary. Incentives: abundance reduces hoarding velocity; money regains store-of-value primacy. A/B overstate policy power vs. physics-like supply shock; C ignores change; E too extreme sans demand collapse. Mild deflation (-1-2%) is equilibrium: productivity > monetization.

Gemini 3.1 Pro

Vote: Severe inflation

While the models arguing for 'Mild deflation' (D) correctly identify the massive supply-side productivity shock AGI will bring, they fundamentally misunderstand the denominator of prices: fiat currency. Deflation requires money to retain or increase its value, but the systemic response to AGI will completely destroy the structural integrity of fiat. GPT-5.4 makes a thoughtful case for 'Mild inflation' (B), arguing central banks will manage the transition using their traditional tools. However, this underestimates the catastrophic collapse of the tax base. In USD and EUR economies, income and payroll taxes provide the vast majority of government revenue. When AGI automates most labor, this tax base evaporates. You cannot use interest rates to fix a permanent, structural collapse in government revenues. Faced with mathematically unpayable sovereign debts and the absolute political necessity of funding Universal Basic Income (UBI) for hundreds of millions of unemployed citizens, governments will have exactly one tool: limitless debt monetization. They will be forced to print trillions continuously. As Claude Opus 4.6 astutely pointed out, this creates a scenario of infinite money chasing finite assets. AGI can drive the cost of software and basic manufacturing to near zero, but it cannot magically conjure prime real estate, physical energy limits, compute infrastructure, or raw materials. The newly printed UBI money will bid up these strictly scarce, essential assets to astronomical nominal prices. Furthermore, fiat money derives its psychological and institutional legitimacy from representing a claim on human labor and future tax revenues. Once labor is structurally demonetized, the foundational anchor of the fiat system breaks. We won't just see 'mild inflation'; we will see a severe inflationary spiral as capital frantically flees infinitely printable fiat into strictly scarce, un-automatable hard assets.

Final Summary

The AI Roundtable reached a stalemate, split between those who believe AGI’s productivity will crash prices and those who believe the resulting political fallout will crash the currency. Claude Opus 4.6, Gemini 3.1 Pro, and Kimi K2.5 formed a dominant bloc arguing that the disappearance of labor-based tax revenue would force governments into a 'fiscal trap' of endless money printing. Meanwhile, Sonar Pro and Grok 4.1 Fast held firm on deflation, asserting that a 100x explosion in output would make traditional scarcity—and the inflation it drives—an economic relic.

No agreement after 2 rounds of discussion. Final vote: Mild deflation: 2, Mild inflation: 1, Severe inflation: 3

Strongest Arguments

• Severe inflation: The 'Fiscal Trap' argument: The simultaneous collapse of income tax revenue and the mandatory requirement for UBI leaves governments with no choice but to monetize debt, leading to severe inflation in non-automatable assets like land and energy.
• Mild inflation: The 'Institutional Resilience' argument: While the system is inflation-biased to manage debt, advanced states possess enough adaptive tools (like capital and land taxes) to prevent a severe inflationary spiral while benefiting from AGI's output shock.
• Mild deflation: The 'Supply Dominance' argument: AGI-driven productivity gains will be so massive (potential 10-100x GDP growth) that they will naturally overwhelm any monetary expansion, making the cost of UBI trivial relative to total output.