Ending extreme poverty could cost $318 billion a year, AI-driven study finds

Ending extreme poverty would cost about $318 billion per year, or roughly 0.3 percent of global GDP, according to a new report by anti-poverty researchers that uses AI tools to produce unusually granular data on the ground. The analysis envisions targeted direct cash transfers that would ensure that virtually everyone has enough to pay for essentials such as food, shelter, and medicines.

The estimate frames the problem in broadly attainable terms. The study uses the World Bank’s extreme poverty line, recently raised to $3 per day, to measure who would be lifted from life-threatening poverty. With this benchmark, the transfers could pull hundreds of millions of people—roughly 8 percent of the world’s population—out of extreme poverty, though many would still live in poverty. The team, led by Paul Niehaus of GiveDirectly, notes that the plan is logistically challenging but not insurmountable. GiveDirectly is a pioneer in no-strings-attached cash transfers that reach people in need directly. The analysis contends that the approach, calibrated with existing government and NGO data, could be implemented at scale with careful oversight. Niehaus and collaborators emphasize that while the cost is nontrivial, it is within reach if political will and coordination align across countries.

What sets this estimate apart is not simply the price tag but how it derives the number. Previous work has often analyzed the so-called poverty gap—the amount needed to push every poor household exactly to the poverty line. Under that approach, a family earning $1 per day would need $1.15, and a neighbor earning $2 would get 15 cents, which yields a very low figure. Critics say this is a thought experiment more than a practical plan because the precise needs of every household simply do not exist in current data. In contrast, the new study uses a broader set of data that governments already track and applies AI-based analysis to estimate transfer amounts. While not perfect, the method is argued to be closer to something that could be implemented in the real world.

What’s different about the new approach helps explain why researchers describe the price tag as surprisingly modest. The authors acknowledge that the plan would be an enormous logistical undertaking, but they contend it is not inherently prohibitively expensive. The real challenge, they say, is governance: turning a new funding obligation into sustained, prioritized action across countries with diverse political incentives.

That insight echoes what researchers describe as the central obstacle to ending extreme poverty: political will. The study notes that wealthier nations have begun deprioritizing affordable anti-poverty programs, a trend with tangible consequences for vulnerable populations. Yet there is a historical counterpoint: over the past several decades, the world has dramatically reduced the share of people living on less than $3 per day, even as the global population grew by billions. Most remaining pockets of extreme poverty cluster in sub-Saharan Africa, where growth has been uneven and demographic momentum is strong. If current patterns continue, the number of extremely poor could still rise in the absence of new, large-scale interventions.

The researchers argue that finishing the job will require someone—whether governments, philanthropists, or a broad coalition of citizens—to underwrite the relatively small price tag and to sustain investments long enough to shift trajectories. In World terms, the implication is clear: historic gains against poverty have required long-term commitments beyond episodic aid. Ending the worst cases of poverty, the study concludes, would demand not just money but a willingness to reorganize priorities so that a modest, continuous flow of cash reaches those in need.

In short, the new analysis presents a case for practicality wrapped in a policy challenge. The proposed price tag is small by many global standards, but the political and institutional work to place cash transfers at scale remains the decisive barrier. If policymakers and supporters can overcome that hurdle, the economists and anti-poverty advocates say, a substantial share of the world’s extreme poverty could be eliminated within a generation.