Anthropic veterans consider buying remote land due to AI risks

Long-serving employees at Anthropic are reportedly considering purchasing land in remote regions of the United States as a contingency plan in the event that artificial intelligence development spirals out of control.

Reports indicate that these early employees are influenced by the ‘Effective Altruism’ subculture and a network of individuals focused on existential risks. Discussions regarding such scenarios have existed since the company's early stages, including hypothetical ‘Manhattan Project-style’ arrangements where researchers might relocate to electromagnetically shielded government facilities in the desert to continue AI development.

This mindset is linked to a broader intellectual movement, including figures like Eliezer Yudkowsky, who has warned about uncontrolled AI for decades. This group previously focused on other catastrophic risks, such as supervolcanoes and asteroid impacts, before shifting focus to AI-related existential threats.