How the Flow of Capital Influences the Course of History: Capital as a Signal for Talent, and What That Means in the Age of Artificial Intelligence

This paper asks a narrow question: does the direction of private capital decide which problems the most capable people work on, and if so, what does that imply for where artificial intelligence will actually land? It proposes a restrained model in which large movements of capital act as a Signal that reallocates talent with a lag, and reads that model against three kinds of evidence. First, capital flows: AI firms took 61 percent of global venture investment in 2025 (USD 258.7 billion of USD 427.1 billion) and roughly 80 percent in the first quarter of 2026, while the four largest United States technology companies guided combined 2026 capital expenditure of about USD 725 billion against a United States federal research budget of USD 192.2 billion and a National Institutes of Health budget of USD 47.2 billion. Second, talent: the share of new North American AI doctorates entering industry rose from 44 percent in 2010 to 66 percent in 2019, although 2022 to 2024 growth in new AI doctorates went to academia, a real qualification the paper reports in full. Third, the long-horizon gap: causal evidence from cancer trials, the shrinking antibacterial pipeline, and the record of enterprise AI pilots all point the same way -- private capital is tilted away from problems with long feedback loops regardless of their social value. Historical cases (wartime R&D, Apollo, the Human Genome Project, AlphaFold) show that tilt can be changed by deliberately funded demand. The paper closes with five policy levers and a specific recommendation for late entrants such as India.