If AI renders human jobs unnecessary, who determines access to food?

The question of how we will be fed is perhaps the most pressing issue being overlooked as discussions about artificial intelligence (AI) and its potential to displace human jobs continue to escalate. Despite the looming threats posed by this technology, most adults of working age have remained gainfully employed throughout history, even as trepidations about job automation have resurfaced since the Industrial Revolution. Yet, we currently find ourselves lacking a robust dialogue about the steps we need to take should AI ultimately claim a significant share of our labor force.

Proponents like Open AI’s Sam Altman affirm that “the future can be vastly better than the present,” positing that the advent of AI will lead to unprecedented wealth. However, such an assertion might only be a pipe dream for the majority of people, excluding Altman and other affluent tech leaders. The assumption of widespread prosperity resulting from AI overlooks the complex challenges of wealth distribution that have always accompanied economic booms.

Even in a scenario where AI generates immense economic wealth, addressing the political ramifications of this wealth distribution becomes essential. At this critical juncture, we require a deep, open discourse regarding how the benefits of this prosperity will be shared amongst all members of society.

The inquiry into how we will be fed hinges on two crucial components. The first involves crafting an effective system for redistributing economic resources as machines take over various roles in the labor market, inevitably driving labor income toward near-zero levels. The second and more significant question pertains to how this economic reorganization will reshape the dynamics of power.

In a potential future where machines are responsible for nearly all economic output, who will wield the authority to dictate taxation policies, especially as labor income evaporates — traditionally a key source of government revenue in many developed nations? Who will control the allocation of resources, determining how much those without any stake in AI ventures will consume?

Consider how society will function in a landscape dominated by machines that generate economic assets while a select handful of techno-billionaires make decisions on the distribution of global resources — from financial assets to energy and minerals. Who will be involved in discussions about whether to prioritize healthcare, agriculture, or education over technological advancement?

United Nations Secretary-General António Guterres expressed an urgent need for “guardrails that preserve human agency, human oversight, and human accountability” during the recent AI Impact Summit in New Delhi. He asserted that the future of AI should not rest in the hands of a privileged few or be left to the whims of a limited set of nations.

In the tech community, the ongoing debate around the “alignment” challenge — ensuring that AI operates in alignment with the interests of its operators — is prevalent. A more pressing concern, however, is aligning the objectives of AI systems and their owners with the overarching goals of society. The implications of AI will affect every individual, yet the existing democratic frameworks appear inadequate to contain the influence of the oligarchs steering these groundbreaking technologies.

Historically, technological advancements have played a vital role in the global spread of democracy, with the emergence of an urban working class necessitating political representation. However, if the work of ordinary individuals becomes superseded by machines, what will remain of their ability to influence governance?

Anton Korinek and Lee Lockwood from the University of Virginia have proposed a foundational overview of how public finance might evolve in the AI age. Initially, they suggest that consumer taxes could help bridge the gap created by the diminishing role of labor income. Yet, as we delve into a world led by artificial superintelligence, the extent of human consumption could decline, necessitating a tax on capital to alleviate the fiscal pressures that will arise.

Additionally, these taxes could potentially be employed to intentionally decelerate the transition to this new economic paradigm. Korinek and Joe Stiglitz from Columbia University advocate for steering tech investments towards enhancing worker productivity rather than replacing human labor altogether, utilizing taxes as a mechanism during transitional phases.

Korinek and Lockwood also recommend alternative taxation strategies, such as taxes on essential resources like land, data, and monopoly profits that yield little social benefit. While these initiatives seem achievable, convincing the holders of disruptive technologies to engage in practices that promote resource sharing presents a significant challenge.

Current tax levels in the United States reflect a mere 26% of GDP, which is 8 percentage points below the average for the OECD, with capital taxes hovering around 2% of GDP. For society to sustain itself as reliance on government assistance increases in a world devoid of traditional wages, these figures must rise significantly.

Given the advocacy for bold reforms, one should not expect rapid change. Despite enthusiasm for the OECD’s global tax agreement finalized in 2021 aimed at curtailing tax evasion among American tech giants like Amazon and Google, political maneuvering continues to complicate progress. The Biden administration initially supported the deal, but it met swift opposition from Donald Trump’s administration, which withdrew support early in 2025.

Innovative solutions may be crucial to sustain society as we anticipate the potential fallout of the AI revolution. One suggestion involves distributing equity from AI enterprises directly to the public. Instead of taxing only monetary returns, a more radical idea might involve the government expropriating a portion of equity to redistribute to the populace, thereby ensuring that Americans gain tangible benefits from AI’s advancements.

According to Korinek and Lockwood, “If AI development stalls, returns remain modest; if AI transforms the economy, returns are likely to rise.” This automatic adjustment may be essential in dealing with the unpredictability surrounding developments in AI.

Yet, these ambitious proposals face formidable obstacles. For substantial change to occur, governments must act decisively before AI surfaces as an insurmountable force — an eventuality that seems increasingly unlikely in today’s political landscape.

Tech moguls leading this revolution have actively resisted oversight and efforts to redistribute wealth. Even with notable figures like Lina Khan striving to curb monopolistic powers during her tenure at the FTC under President Biden, significant change proved elusive.

Meanwhile, Silicon Valley elites continue to pour financial resources into steering political discourse. As a contingency, they are exploring the establishment of “network-states,” be it in regions like Greenland, Honduras, or the Caribbean, with the aim of circumventing democratic governance if their needs cannot be met within existing political structures.

It raises an unsettling question: in a scenario where AI eventually displaces all human labor, what strategies will we have left to ensure our basic needs are met? In a world dominated by immense technological powers, the only option may become one of subservience, whereby we must humbly approach wealthy patrons for our sustenance.