Anthropic Doubles AI Policy Donation to $40M for 2026 Midterms

The AI money in Washington is starting to look less like lobbying and more like full campaign infrastructure. Anthropic said it will put another $20 million into Public First Action, the nonprofit advocating for AI safeguards, doubling its total commitment to the group to $40 million. The first tranche went in February 2026, and the new one lands as the AI industry's political spending is visibly accelerating into the 2026 midterms.

The framing matters. Anthropic told The Hill this money is going to Public First Action's 'public-education and policy mission' and 'cannot be used' to influence federal, state or local candidates. That is the 501(c)(4) side of a two-part vehicle. The related super PAC, Public First, is where the campaign spending actually happens. According to reporting summarized by CNBC, Public First Action was founded in 2025 by former Representatives Brad Carson, a Democrat, and Chris Stewart, a Republican, and it plans to back 30 to 50 candidates from both parties in state and federal races.

The company check is not the only Anthropic-linked money on the table. Campaign finance filings released the week before showed CEO Dario Amodei gave $1 million to the super PAC, his first seven-figure political donation, and Anthropic employees gave a combined $2.15 million over the last quarter. On the other side of the argument sits Leading the Future, a super PAC backed by OpenAI President Greg Brockman and Andreessen Horowitz, which reportedly had $31 million on hand at the end of June.

The honest caveat is that this is disclosed money in a single announcement window, and the Hill piece does not spell out which specific bills or races Anthropic wants the fresh $20 million to influence. The legal wall between 501(c)(4) education work and affiliated super PAC candidate spending is a distinction reasonable people can read skeptically. The reporting also does not resolve whether other frontier labs will match this pace or sit it out.

What is worth watching is straightforward. AI regulation in 2026 is no longer being settled by hearings and think-tank op-eds. It is being argued inside the same primaries and general elections that decide everything else, and the industry has now signaled it is willing to fund both sides of that argument at real scale.