Tax High-Frequency AI Trading to Fund Literacy

This proposal introduces a small excise tax on stock trades executed by AI algorithms operating at speeds and volumes beyond the capacity of human traders. The revenue generated would be used to fund a national digital literacy program and help the SEC monitor algorithmic market manipulation that has become increasingly difficult to detect. The proposed tax rate is 0.02 percent per transaction, which advocates argue is negligible for any single trade but would generate an estimated $6 billion to $9 billion annually given the volume of algorithmic trading, which now accounts for roughly 70 percent of all U.S. equity market transactions. The digital literacy program would distribute funds to states through block grants, with priority given to rural school districts and underserved urban communities where access to technology education is most limited. Approximately 15 percent of the revenue would be allocated directly to the SEC for hiring data scientists and building real-time surveillance infrastructure capable of identifying AI-driven market manipulation patterns. The financial industry, represented by groups such as the Securities Industry and Financial Markets Association, opposes the tax on the grounds that it would reduce market liquidity, widen bid-ask spreads, and ultimately increase costs for retail investors. Proponents counter that the tax is too small to meaningfully affect legitimate trading but large enough to discourage the most aggressive high-frequency strategies that contribute to flash crashes and market instability. Similar transaction taxes exist in the United Kingdom, France, and Italy, where studies have shown minimal long-term impact on market depth. The proposal includes an annual review mechanism requiring the Treasury Department to assess whether the tax rate should be adjusted based on market conditions and revenue outcomes. Implementation would be phased over two years, beginning with equities and expanding to options and futures in the second year.