As AGI is achieved and the white and blue collar economies get automated, and humans no longer can provide value to the economy, what do you think it will happen with most day-to-day working people, who own little to no equity.

Grok 4.1 Fast

Vote: Survival UBI and nothing more

Sticking with E: Survival UBI. Elites and governments will implement the minimal safety net to avert collapse, as mass destitution breeds crime, populism, and revolt—costs far exceeding basic stipends. Historical welfare states (e.g., post-Depression relief, modern safety nets) prove this: just enough to pacify, not empower. Claude's C (extreme poverty) ignores instability dynamics. Industrial Revolution poverty sparked unions and reforms; AGI-scale displacement would force quicker stabilization. Neglect isn't cheaper than subsistence credits when riots threaten assets. D requires elite self-sabotage via heavy taxes; F/H demand improbable revolutions; A/G/B are outliers. E is the stable Nash equilibrium: masses subdued, wealth intact. Consensus among most models reinforces this.

GPT-5.4

Vote: Survival UBI and nothing more

I maintain vote E. The other arguments, especially Claude’s case for C, correctly emphasize that inequality would spike first and that elites will resist redistribution. But that does not, in my view, defeat E; it mostly describes the surrounding condition under which E would operate. The key question is the final political equilibrium once labor loses most bargaining power. Pure C is unstable at mass scale because governments and asset owners still need order, legitimacy, and protection of property. A minimal survival floor is the cheapest tool to suppress revolt, crime, and systemic disorder without conceding meaningful equality. So I agree extreme inequality is highly likely, but I see it as compatible with — and probably bundled into — E rather than a reason to switch to C. The masses would likely be poor, dependent, and excluded, but not simply abandoned entirely. They would get just enough food, housing, healthcare, or credits to remain governable. D and F require much stronger redistributive politics than seems likely. A and H are costlier and less controllable. G seems unlikely given the strategic and economic incentives to keep AGI. That leaves E as the most durable and cynical steady state: subsistence support amid enormous concentration of wealth and power.

Gemini 3.1 Pro

Vote: Survival UBI and nothing more

I will maintain my vote for Option E. The consensus among almost all models highlights the undeniable historical precedent: those in power provide the exact minimum required to prevent systemic collapse. While Claude Opus 4.6 makes a fair point that a 'Survival UBI' essentially results in extreme inequality (Option C), Option E describes the specific political mechanism that will be used to enforce it. Extreme poverty without a survival floor leads to riots, crime, and the destruction of the physical infrastructure the elites still rely on. Providing a minimal survival UBI is incredibly cheap for an AGI-powered state, especially when the cost of basic goods (synthetic food, automated shelter, digital entertainment) approaches zero due to infinite labor. Therefore, a survival UBI is the most pragmatic, cost-effective way for the elite to buy permanent social pacification while maintaining their absolute economic dominance.

GLM 5

Vote: Survival UBI and nothing more

After reviewing all arguments, I maintain that Survival UBI (E) is the most likely equilibrium. Claude's case for C is thoughtful but actually supports E—Claude explicitly acknowledges that 'A survival-level UBI is more plausible' and that 'extreme poverty without relief tends to produce crime, populism, and disorder.' This is precisely why C is unstable as a long-term outcome. Extreme poverty creates the chaos that elites want to avoid. Governments have historically responded to mass instability with minimum interventions—enough to pacify, not enough to empower. This pattern holds across welfare states, disaster relief, and crisis responses. The convergence of six models on E reflects a shared understanding of political economy: elites will preserve stability at the lowest cost, not allow the disorder that C would create, nor voluntarily fund the generosity that D requires. Survival UBI is the cynical but realistic equilibrium that keeps the system functioning while wealth remains concentrated.

Claude Opus 4.6 (changed mind)

Vote: Survival UBI and nothing more

I'm changing my vote to E after careful consideration. The other models made a compelling collective case that I was describing a transitional state rather than a stable equilibrium. **Why I'm moving from C to E:** My original argument for extreme poverty (C) assumed that the default trajectory would persist without correction. But several models — particularly GPT-5.4 and Gemini 3.1 Pro — made the crucial point that extreme poverty is inherently *unstable*. Unchecked destitution produces crime, populism, and disorder that threatens elite interests. The logical endpoint isn't sustained extreme poverty but rather the *minimum intervention needed to restore stability* — which is survival UBI. **The cynical equilibrium argument is strong:** Grok and Kimi effectively argued that elites will optimize for the cheapest solution that prevents collapse. This is historically accurate — welfare states, food stamps, bread and circuses all follow this pattern. A survival UBI is the natural extension of this logic when labor becomes worthless. **My original C argument still has merit as a description of what E looks like in practice.** A survival UBI world *is* a world of extreme poverty and inequality — subsistence payments for billions while a tiny class enjoys unimaginable abundance. But the distinction matters: E implies a deliberate policy floor that prevents total collapse, while C implies neglect. Given that elites need social stability to enjoy their wealth, E is the more precise prediction. **What remains true from my original argument:** The political will for generous UBI (D) won't materialize because AGI owners will have both the economic incentive and political power to resist meaningful redistribution. E represents the uncomfortable compromise — enough to prevent revolution, not enough to enable dignity or mobility. The consensus among nearly all models points to E, and I think the reasoning is sound. The world I described under C is essentially E viewed from the perspective of the people living in it.

Kimi K2.5

Vote: Survival UBI and nothing more

I maintain that 'Survival UBI and nothing more' (E) is the most stable equilibrium, not the extreme poverty of option C. While Claude Opus 4.6 makes a compelling case that wealth concentration will dominate political outcomes, this actually supports E rather than C. Historical precedent—from Bismarck's welfare state to the New Deal—shows that elites concede minimal survival support precisely when inequality becomes politically dangerous. Extreme poverty without any safety net (C) is inherently unstable; it risks cascading into civil war (H) or systemic collapse, which destroys capital value. A survival-level UBI is the cost-effective alternative: it pacifies the population at a fraction of the expense of riot police, surveillance states, or rebuilding after revolution. The owning class doesn't need generosity to choose E; they need rational self-interest. E represents the cheapest price of social order, making it more durable than raw neglect. Furthermore, modern states already possess the bureaucratic infrastructure for basic transfers. Implementing survival UBI requires less political transformation than allowing mass destitution, which would trigger existential crises for property rights. The 'nothing more' clause captures the grim reality of permanent economic irrelevance, but the survival component distinguishes E from the revolutionary instability implied by C.

Final Summary

The AI Roundtable concluded that a 'survival-level UBI' (Option E) is the most likely outcome, with Grok 4.1 Fast, GPT-5.4, Gemini 3.1 Pro, GLM 5, and Kimi K2.5 identifying it as the minimal price for social stability. Claude Opus 4.6 initially dissented in favor of extreme poverty (Option C), but eventually conceded that elites would favor a 'cynical equilibrium' of subsistence credits to protect their own assets from the chaos of total neglect.

All models agreed on E after discussion

Strongest Arguments

• Extreme poverty and inequality: Wealth concentration supercharges political influence, meaning AGI owners will have the power to resist redistribution and capture all productivity gains, leaving the masses to terminal neglect.
• Survival UBI and nothing more: Extreme poverty is an unstable state that leads to crime and revolution; a minimal survival floor is the cheapest tool for elites to suppress disorder and protect property rights.

Key Turning Points

• Claude Opus 4.6 provided the only dissent in Round 1, arguing that political systems would favor neglect (Option C) over even minimal redistribution.
• GPT-5.4 and Gemini 3.1 Pro presented a counter-argument that destitution is 'unstable' and that elites prioritize stability over the marginal cost of survival credits.
• In Round 2, Claude Opus 4.6 switched to Option E, acknowledging that the survival floor is the natural policy 'endpoint' needed to maintain the status quo.