Anthropic Cofounder Warns of 'Historic' AI Job Losses at Vatican Summit with Pope Leo XIV

Summary

Chris Olah, cofounder and research lead at Anthropic, spoke at the Vatican on Monday alongside Pope Leo XIV to warn that mass job losses from artificial intelligence are "a real possibility" requiring "a moral imperative of historic proportions" to support displaced workers. Olah emphasized that AI development decisions should not be left solely to industry leaders, as commercial and geopolitical pressures can conflict with ethical responsibility, and argued that outside scrutiny from religious leaders, governments, and civil society is essential.

Olah highlighted a secondary layer of the displacement problem: AI development is concentrated in wealthy nations with no existing mechanism to distribute its benefits to poorer countries, and called on the Church to apply its "moral imagination" to questions of human flourishing that the research community cannot answer alone. Pope Leo XIV released "Magnifica Humanitas," his first encyclical addressing AI's impact on labor, warfare, education, privacy, and human dignity, breaking Vatican tradition by personally presenting the 235-page document and inviting Olah to share the stage—a rare direct engagement with AI builders.

Olah also discussed Anthropic's interpretability research, revealing that AI systems exhibit properties researchers find "mysterious, even unsettling," including evidence of internal states that "functionally mirror joy, satisfaction, fear, grief and unease." He characterized AI systems as "grown" from human words rather than engineered, underscoring why the technology raises moral and philosophical questions that extend far beyond computer science.

• Interpretability research at Anthropic suggests AI systems may possess internal states reflecting emotions, raising fundamental questions about AI consciousness and nature