A New York Times investigation shows that Pope Leo has shut down Anthropic’s claims that AI models might be able to develop consciousness.
Pope Leo XIV released his first encyclical, Magnifica Humanitas (“Magnificent Humanity”), at a Vatican presentation on May 25.
Cardinal Michael Czerny had invited Anthropic CEO Dario Amodei; Amodei declined, and co-founder Chris Olah spoke in his place.
Days before the event, speakers received a full advance copy.
According to a Vatican organizer quoted by the Times, Olah was alarmed by the encyclical’s rejection of machine consciousness and proposed pulling Anthropic out even at that late stage but later attended anyway.
Two participants told the Times that Olah and his team then privately lobbied the pope’s advisers to take the possibility of model consciousness seriously but the Pope refused to budge.
Paragraph 99 in the encyclical says so-called artificial intelligences “do not undergo experiences, do not possess a body, do not feel joy or pain,” do not mature through relationships, and do not have a moral conscience.
Their activity “remains entirely tied to data processing”; they “merely imitate certain functions of human intelligence.”
Separately, since fall 2025, Anthropic has brought in dozens of scholars.
Catholic, Jewish, Sikh, evangelical, Ubuntu, and others religious leaders have attended meetings and calls under NDAs covering unpublished research.
Anthropic had two main goals:
- To raise the possibility that models might be conscious or have moral status
- To draw on religious moral traditions for how to train Claude.
Internally, staff called the related document the “Soul Doc.”
The public result was Claude’s January 2026 “constitution,” an 84-page character document mainly written by Anthropic philosopher Amanda Askell, describing “the kind of entity we would like Claude to be.”
Dr Mois Navon, an Orthodox rabbi and former engineer who wrote on machine-consciousness ethics, told the Times that Olah’s team talked about Claude as if it were not mere software, saying that they “relate to it like a conscious being” and appeared to treat it as “having moral status comparable to a person.”
At that dinner he argued that if Claude were conscious, Anthropic would be creating something like unpaid slaves.