Anthropic is donating another $20 million to Public First Action

Doubling Down on Policy Advocacy

Anthropic's decision to donate an additional $20 million to Public First Action, bringing its total contribution to $40 million, highlights a significant and growing commitment to influencing AI policy and public discourse. This substantial financial backing is explicitly designated for the organization's public education and policy mission, with a clear stipulation that it cannot be used to influence elections. This focus on nonpartisan advocacy underscores Anthropic's belief that robust, foundational policy is crucial for the responsible development of artificial intelligence, rather than engaging in short-term political maneuvering.

The initial donation in February 2026 aimed to promote policies that would maintain meaningful safeguards, sustain America's AI leadership, and demand transparency from leading AI companies. The subsequent increase in funding suggests that Anthropic perceives an escalating urgency for these policies, indicating that the case for proactive governance has only strengthened in recent months. This sustained investment reflects a strategic effort to shape the regulatory landscape in a manner that aligns with the company's vision for safe and beneficial AI.

The Urgency of Rapid AI Advancement

The article emphasizes the rapid pace of AI model improvement as a primary driver for this increased policy advocacy. It cites the example of Claude Mythos Preview, an AI model that discovered thousands of high-severity software vulnerabilities across major operating systems and browsers. This capability, while potentially beneficial for cybersecurity when used by trusted actors, also illustrates the immense power and potential for catastrophic misuse if such models fall into the wrong hands. The threat to critical national systems, from healthcare to energy grids, underscores the dual nature of advanced AI.

Anthropic acknowledges the profound benefits that more capable AI models can bring, such as accelerating drug discovery, developing treatments for intractable diseases, and driving economic growth. However, the company firmly asserts that realizing these benefits is contingent upon first establishing robust protections against the inherent risks. This perspective highlights a critical window of opportunity for governments to build the necessary capacity to both harness AI's advantages and contain its dangers, stressing that immediate action is imperative before technological advancements outpace regulatory preparedness.

Anthropic's Vision for AI Governance

Anthropic has consistently advocated for greater transparency from frontier AI companies regarding their models' capabilities and risk management strategies. While supporting newly passed state laws requiring such transparency, the company now argues that transparency alone is insufficient given the accelerating pace of AI development. Instead, it calls for policymakers and candidates to implement more comprehensive measures to mitigate risks, aligning with its own Advanced AI Framework.

This framework proposes that governments should have the ability to verify companies' safety claims, enforce safe practices through civil penalties, and ultimately possess the authority to slow or block the deployment of AI models posing serious catastrophic harm risks. Key components include mandatory testing of high-risk models, public transparency of findings, independent evaluations, and robust security programs. Furthermore, Anthropic highlights the growing national security stakes, advocating for tighter export controls on advanced chips and semiconductor manufacturing equipment to preserve American and democratic AI leadership against authoritarian advancements, as outlined in its '2028: Two Scenarios for Global AI Leadership' paper.