AI Job Displacement Revives Universal Basic Income Debate

The rise of artificial intelligence has reignited debate over how the United States should protect workers from technological displacement. While universal basic income proposals have gained traction among tech leaders and some politicians, the question remains whether such radical departures from existing policy make sense—or whether the current welfare state offers a more practical foundation for expansion.

Andrew Yang's 2020 presidential campaign brought UBI into mainstream political discourse with his "Freedom Dividend" proposal: $1,000 monthly payments to every American adult, funded in part by eliminating existing benefit programs like SNAP, SSI, TANF, and WIC. Yang framed the policy as a necessary response to automation, particularly self-driving vehicles that would displace truckers and related service workers.

The AI boom has only intensified these conversations. Major AI company leaders have endorsed various redistributive proposals, from UBI to partial public ownership to new taxation schemes. Their rhetoric often treats current welfare infrastructure as outdated technology—analog systems inadequate for a transformed economy.

Why it matters

The framing of this debate will shape how millions of Americans access economic security in the coming decades. If policymakers treat AI as a complete rupture requiring entirely new systems, they risk dismantling programs that currently keep vulnerable populations afloat. Alternatively, expanding and modernizing existing infrastructure could provide faster, more politically durable protection.

The case against starting from scratch

Policy experts who have spent careers defending social programs view UBI proposals with deep skepticism. Yang's Freedom Dividend would have forced recipients to choose between existing benefits and the monthly check—a trade-off that would likely harm those who depend most on targeted assistance.

These programs, while imperfect, represent decades of political coalition-building and administrative development. Social Security's success came from tying benefits to payroll contributions, creating a sense of earned entitlement. The Earned Income Tax Credit, despite its limitations and racial biases in practice, has survived multiple hostile administrations precisely because it's embedded in the tax code and tied to work.

According to reporting from Dissent Magazine, 4.5 million Americans lost food stamp benefits in the past year alone, including 1.5 million children. Existing programs remain under constant threat—adding an opt-out mechanism could accelerate their collapse.

Building on existing infrastructure

Rather than replacing the welfare state, several strategies could strengthen it for an AI-disrupted economy. These include decoupling health insurance and other benefits from employers, rebuilding Medicaid and SNAP, expanding unemployment insurance, and reforming TANF to ensure funds reach families.

The 2021 Child Tax Credit expansion demonstrated both the potential and fragility of new approaches—it cut child poverty by one-third but proved politically vulnerable. The EITC has shown greater durability, though it becomes less effective if formal employment disappears.

New proposals are emerging that work within existing frameworks while addressing AI-specific challenges. Congressional Progressive Caucus Chair Greg Casar has proposed a token tax tied to unemployment rates. Other lawmakers are exploring traditional wealth and corporate profit taxes to capture AI-generated gains.

The path forward

Technological change has historically opened political space for policy innovation—from Reconstruction-era railroad speculation to post-atomic national security infrastructure. AI may similarly enable previously unthinkable reforms.

The challenge is harnessing that possibility without abandoning hard-won protections. A jobs guarantee including care economy investments, expanded public goods, and strengthened safety net programs could address AI displacement while building on proven administrative capacity.

These details were first reported by Marc Aidinoff, an assistant professor of the history of technology at Harvard University, writing in Dissent Magazine.