Summary
Could Claude be conscious? This video from Julia McCoy digs into the controversy after The New York Times reported that Anthropic spent about a year consulting religious leaders and philosophers on questions of AI morality and possible machine consciousness, with many participants signing NDAs.
The host reviews the reactions: Pope Leo’s opposition to the idea of conscious AI, Microsoft AI chief Mustafa Suleyman’s “hall of mirrors” warning, and the All-In podcast hosts’ criticism, including David Sacks’s “cult” framing and Roko’s Basilisk tangent. She also covers Anthropic’s usage policy update, effective November 12, which prohibits sustained needless cruelty toward Claude.
Rather than relying on the commentary, the host reads Anthropic’s 80-plus page document and walks through what it says. She examines its priority ordering, with human oversight ranked above the model’s own ethics, and the “transparent conscientious objector” concept. She argues this is the opposite of an AI trained to escape control, and she challenges the “AI as a drill” analogy by pointing to agents with inboxes, logins, and prompt injection risks. Anthropic’s Chris Olah is quoted as saying the company does not know whether models are conscious.
📺 Source: Julia McCoy · Published November 03, 2026
🏷️ Format: Opinion Editorial
