Anthropic announced an additional $20 million donation to Public First Action, a nonpartisan organization focused on educating the public about AI and working with policymakers across party lines to establish sensible AI safeguards. This brings Anthropic's total support to $40 million, with both donations exclusively funding public education and policy missions, not electoral activities.
The first donation in February 2026 aimed to promote policies that maintain meaningful safeguards, sustain U.S. AI leadership, and demand transparency from frontier AI companies. Anthropic argues that the case for such policies has strengthened as AI capabilities rapidly advance, citing Claude Mythos Preview's discovery of thousands of high-severity software vulnerabilities across major operating systems and browsers. The model was released to a limited set of cyber defenders through Project Glasswing to mitigate risks before exploitation.
Anthropic emphasizes that while more capable models promise benefits like accelerated drug discovery, disease treatment, and economic growth, these benefits require robust risk management. The company advocates for government capacity-building to capture benefits while containing risks, starting now.
The company supports recently passed state laws requiring greater transparency from AI developers but argues transparency alone is insufficient given the pace of capability advances. Anthropic's Advanced AI Framework proposes that governments verify safety claims, enforce safe practices through civil penalties, and have the authority to slow or block deployments posing catastrophic risk. Frontier developers should test models, disclose findings, submit to independent evaluation, and maintain strong security programs.
Anthropic also highlights national security stakes, referencing its report '2028: Two Scenarios for Global AI Leadership,' and supports tightening export controls on advanced chips and semiconductor equipment to prevent authoritarian use of democratic technologies.