Anthropic Commits $40M to AI Policy Group Ahead of Midterms

Anthropic has doubled down on its political advocacy efforts, announcing a second $20 million donation to Public First Action, bringing its total commitment to the nonprofit organization to $40 million ahead of November's midterm elections.

The AI safety company said the funding reflects growing urgency around AI policy as more advanced models like Claude Mythos enter development. "We need policymakers and candidates to put forward measures that mitigate risks," Anthropic stated Wednesday.

How the funding flows

Public First Action operates as a nonprofit organization that isn't required to disclose its donors, unlike traditional super PACs. The group has raised $80 million to date and maintains ties to three allied super PACs that have backed both Republican and Democratic candidates across various races.

While Anthropic says its donation "cannot be used to influence the election of any candidate for federal, state, or local office," the nonprofit has directed funds toward advertising that spotlights specific lawmakers. Its first advertisement highlighted Republican Sen. Marsha Blackburn's work on AI and child online safety legislation as she campaigns for Tennessee governor.

Former Republican Congressman Chris Stewart of Utah and former Democratic Congressman Brad Carson of Oklahoma cofounded Public First Action. The pair also run separate allied super PACs—Defending Our Values PAC and Jobs and Democracy PAC—that support candidates from their respective parties.

A proxy battle over AI regulation

Public First Action has positioned itself in direct opposition to Leading the Future, a super PAC that has raised more than $75 million from tech industry figures including OpenAI President Greg Brockman, Marc Andreessen, and Ben Horowitz. OpenAI has clarified that Brockman is acting in a personal capacity.

The competing organizations represent diverging visions for AI oversight. Anthropic CEO Dario Amodei has emerged as the industry's most vocal advocate for stringent regulation, proposing last month that the government establish an agency similar to the Federal Aviation Administration. Under his proposal, this body would mandate testing of advanced AI models and have authority to block releases that fail safety evaluations.

The Trump administration and other industry players have warned against such mandatory review processes.

In May, Amodei personally donated $1 million to Public First, the only allied super PAC that supports candidates from both parties.

Why it matters

Anthropicʼs $40 million commitment represents one of the largest corporate investments in shaping AI policy through electoral politics. As Congress considers various AI regulatory frameworks, the company is betting that supporting candidates who favor oversight will prove more effective than lobbying alone. The approach signals that debates over AI governance—once confined to tech circles and think tanks—have become central to electoral strategy, with competing factions of the industry now funding opposite sides of the political landscape.

These details were first reported by Business Insider.