Proposed US legislation would compel certain Artificial Intelligence (AI) companies to transfer a 50 per cent equity stake to a government-managed sovereign wealth fund, according to a new Congressional Research Service (CRS) report released in August 2025. The non-partisan agency cautioned that the measure could face significant constitutional challenges if enacted.

What the Proposed Legislation Says

The American AI Sovereign Wealth Fund Act would require covered companies to make a one-time transfer of half their equity to the fund, which would be overseen by a seven-member commission. As the value of those holdings grew, the fund would distribute direct payments to the American public. The arrangement would also grant the federal government voting rights, board representation, and other corporate governance powers over affected companies.

Constitutional and Legal Concerns

The CRS report noted that companies or shareholders subjected to the compulsory transfer could challenge its constitutionality if the bill became law. The agency, which provides non-partisan research to Congress, did not endorse or oppose the legislation but flagged the legal uncertainty as a material risk. The report comes amid broader discussions in Washington about whether the public should receive a share of the wealth being generated by AI and the infrastructure supporting it.

Trump's Reported Discussions and Scale of AI Investment

President Donald Trump has reportedly discussed possible government equity stakes in private AI companies with industry representatives, according to the report. No details of those discussions had been publicly released as of July 2025. The scale of investment at stake is substantial: AI-related spending is projected to exceed USD 1 trillion globally during 2026, with nearly USD 600 billion concentrated in the United States. Amazon, Google, Meta, and Microsoft alone spent an estimated USD 420 billion on AI infrastructure during 2025. One investment manager cited in the report estimated that additional global AI capital expenditure between 2026 and 2030 could reach approximately USD 7.5 trillion.

Market Concentration and Systemic Risk

Several AI-linked technology companies now account for more than 40 per cent of the total market capitalisation of the S&P 500. These include Nvidia, Alphabet, Apple, Microsoft, Amazon, Broadcom, Meta, and Micron. The report warned that a major fall in their share prices could reduce household and investor wealth and potentially spread stress across the broader financial system. It also examined concerns that large-scale AI-driven job losses could reduce household consumption and increase defaults on mortgages and other financial instruments — though it noted there was no evidence as of July of widespread labour-market disruption attributable to AI.

Precedents and Alternative Models

The CRS report cited historical parallels, including the canal investment boom of the 1830s, British railway expansion in the 1840s, the speculative surge of the 1920s, and the dot-com boom of the late 1990s — each followed by sharp market corrections with broader economic consequences. Possible mechanisms for distributing AI-generated wealth, beyond sovereign equity funds, include universal basic income, tax reforms, and direct public ownership. The report noted that OpenAI and Anthropic have released policy frameworks supporting broader prosperity-sharing from AI. The federal government already holds equity stakes in strategically important industries: the report cited an USD 8.9 billion government equity investment in Intel, investments in quantum-computing and semiconductor firms, and a 'golden share' with special governance rights obtained as a condition for approving Nippon Steel's acquisition of US Steel.

Whether the American AI Sovereign Wealth Fund Act advances through Congress remains to be seen, but the CRS report signals that the debate over who benefits from the AI boom has moved well beyond think tanks and into formal legislative scrutiny.