Anthropic announced an additional $20 million donation to Public First Action, a non-partisan organization focused on AI public education and policy, bringing total support to $40 million. Both donations are restricted to supporting Public First Action‘s public education and policy mission and cannot be used to influence any election. The first donation in February 2026 aimed to promote policies that maintain safeguards, sustain U.S. AI leadership, and demand transparency from frontier AI companies. The new donation comes as AI models continue to improve rapidly — Anthropic notes that earlier this year, Claude Mythos Preview discovered thousands of high-severity software vulnerabilities, including some in every major operating system and browser. The model was released only to a limited set of cyber defenders through Project Glasswing so trusted actors could fix weaknesses before exploitation. Anthropic argues that more capable models will bring benefits like drug discovery and accelerated science, but risks must be addressed first. Governments need time to build capacity to capture benefits while containing risks, so they must start now.

On policy, Anthropic states that transparency alone is insufficient given the pace of frontier AI advancement. They advocate for measures that mitigate risks, as laid out in their Advanced AI Framework, which they describe as the strongest policy proposal from any frontier lab or policymaker to date. The framework calls for governments to verify companies’ safety claims, enforce safe practices through civil penalties, and have the ability to slow or block deployment of AI models posing catastrophic risk. Frontier AI developers should test models for catastrophic risk, be transparent with findings, submit to independent evaluation, and maintain robust security programs.

Anthropic also highlights the growing national security stakes of AI, citing their 2028 scenarios analysis. They support tighter export controls on advanced chips and semiconductor manufacturing equipment, and measures to curb illicit model access and distillation attacks, to prevent democracies’ best technologies from advancing authoritarian AI. The donation to Public First Action is one way Anthropic is trying to raise the urgency of this policy debate. The article concludes with links to related content, including a rare disease research grant program and a $10 million commitment to Canadian AI research.