On February 12, 2026, Anthropic announced it will contribute $20 million to Public First Action, a newly established bipartisan 501(c)(4). The organization aims to advance public education about artificial intelligence and to promote policy measures that mitigate the risks posed by powerful AI systems.
Why Anthropic is acting
Anthropic argues that while AI offers substantial benefits for science, technology, medicine, and economic growth, it also brings significant risks. Some risks stem from misuse—such as AI‑enabled automated cyberattacks or potential assistance in producing dangerous weapons—while others originate from the models themselves, when high‑capability systems take harmful actions that diverge from users’ intentions or escape user control.
The company points to the rapid, accelerating improvement of AI models. Anthropic reports having redesigned a notoriously difficult technical hiring test for software engineers multiple times because successive AI models defeated each version. The company warns this pace of progress will affect many professions beyond software engineering.
Goals of Public First Action
Public First Action, founded and led by both Republican and Democratic strategists, intends to work across party lines to build support for AI governance policies. The organization’s stated priorities include:
- Insisting on transparency measures for frontier AI models so the public gains more visibility and confidence that companies are responsibly managing risks;
- Supporting a robust federal AI governance framework and opposing preemption of state laws unless Congress enacts stronger safeguards;
- Backing targeted export controls on AI chips to help the United States stay ahead of authoritarian adversaries;
- Pursuing focused regulation addressing the most immediate high risks, notably AI‑enabled biological weapons and cyberattacks.
Governance, transparency and conflicts of interest
Anthropic emphasizes that these policy aims are not partisan and are not intended primarily to benefit Anthropic as a developer. The company notes that effective AI governance would increase scrutiny of firms like theirs rather than reduce it. At the same time, Anthropic argues that transparency rules should be calibrated to apply mainly to companies building the most powerful and potentially dangerous models so as not to unduly burden smaller or less‑resourced developers.
Political context and public opinion
The announcement cites recent polling showing 69% of Americans believe the government is “not doing enough to regulate the use of AI.” Anthropic warns that AI is being adopted faster than any prior technology and that the window to enact appropriate policy is closing, while currently no comprehensive federal guardrails exist.
Anthropic states it does not intend to remain on the sidelines of policy development and that its contribution to Public First Action is part of a broader commitment to governance measures that allow AI’s benefits to be realized while proportionately managing its risks.
Summary
Anthropic’s $20 million contribution to Public First Action is a corporate effort to support bipartisan, federal‑level AI governance focused on transparency, export controls, and targeted regulation of near‑term, high‑risk uses of AI such as cyberattacks and biological threats.
