Inviting hard questions

Anthropic is launching a new initiative to gather and address public questions regarding the development and impact of artificial intelligence. While many people recognize the potential for AI to improve scientific progress, medical research, and daily productivity, there are widespread concerns about job displacement, the devaluation of creative work, and the loss of human agency.

To better understand these perspectives, the company has already surveyed over 130,000 people across various countries and demographics. These efforts include the Anthropic Public Record, which polled 52,000 Americans, and the Anthropic Interviewer, which reached 81,000 Claude users globally. The company also utilizes the Anthropic Institute to research societal challenges and relies on a Long-Term Benefit Trust to provide independent oversight of its mission.

As a Public Benefit Corporation, Anthropic aims to align its AI models with beneficial goals while mitigating risks like misuse. Through this new initiative, the company is inviting the public to submit their most difficult questions about the future of AI. In response, Anthropic commits to publicly tracking its progress, reporting on the specific actions taken to address these concerns, and being transparent about where its efforts may fall short of its stated goals.