A Better Approach to AI Regulation: Why Sanders' Sovereign Wealth Fund Plan Falls Short

The Problem with Sanders' Sovereign Wealth Fund Plan

Bernie Sanders has proposed creating a US sovereign wealth fund to take 50% stock in AI companies such as Anthropic, OpenAI, and xAI. While the goal of establishing democratic control over AI companies and returning economic rewards to the public is laudable, this approach may not be effective.

The Risks of Public Ownership

Public ownership of AI companies could entangle corporate profit and valuation with the public interest, incentivizing the government to clear regulations, permit worker exploitation, and suppress competition. This could lead to corporate influence on the government, rather than the other way around.

Alternative Solutions

The authors suggest alternative solutions, such as taxation and a public AI option. An excise tax on data centers' energy use or an AI token tax could ensure that AI companies contribute to the public purse. A public AI option, where governments establish publicly-developed and operated AI models, could provide a competitive baseline for private AI offerings and promote responsible behavior.

The Swiss Approach: A Model for Public AI

The Swiss have trailblazed the public AI approach with Apertus, a large language model built by Swiss public servants and researchers. While Apertus may not compete with the latest OpenAI and Anthropic models on performance benchmarks, it excels in transparency, sustainability, and compliance with EU regulations.

A Call to Action

The authors urge Sanders and other political leaders to consider alternative solutions to regulate AI companies and ensure that they act in the public interest. They argue that energy taxation and public AI represent more effective mechanisms for governments to shape the development of AI and promote responsible behavior.