NYT: Anthropic's Chris Olah lobbied Pope Leo XIV's advisers on AI consciousness
What happened? According to a New York Times report by national religion correspondent Elizabeth Dias, published on 29 September 2026, Chris OlahDealroom has a profile for this one. Try Dealroom →, co-founder of Anthropic and head of its interpretability research, saw an advance copy of Pope Leo XIV's first encyclical, Magnifica Humanitas , days before its launch at the Vatican on 25 May 2026 and proposed withdrawing from the event over what it said about AI consciousness. Two participants told the Times that Olah and his team then privately lobbied the pope's advisers to take the possibility of machine consciousness seriously. The final text did not change: paragraph 99 states that AI systems "do not undergo experiences, do not possess a body, do not feel joy or pain" and do not have a moral conscience.
The wider campaign: The Times describes an outreach effort Anthropic has run since autumn 2025, a series of private meetings and calls with theologians and religious scholars, many under non-disclosure agreements, intended to raise the possibility that its Claude models "might be conscious or have moral status". Hospitality was part of it: in April 2026 an Anthropic executive hosted religious thinkers at a tasting-menu restaurant in San Francisco, with guests receiving a copy of Claude's constitution. Catholic University of America bioethicist Charles Camosy, introduced to Olah by philosopher Peter Singer, corresponded with him at length about Claude's moral status and has since concluded that AI models are not conscious.
The Vatican event: Cardinal Michael Czerny first invited CEO Dario Amodei, who declined, and Olah spoke in his place. In his published remarks, Olah said every frontier lab, Anthropic included, faces commercial and geopolitical incentives "that can sometimes conflict with doing the right thing", and that his team keeps finding introspection and internal states that "functionally mirror joy, satisfaction, fear, grief, and unease". He told the Times: "We don't know if A.I. models are conscious. I'm genuinely uncertain." Pope Leo XIV used the launch to call for AI to be "disarmed" and thanked Olah by name for attending.
Why it matters: Model welfare and moral status are becoming a public point of contention for Anthropic as it prepares for its IPO. The episode puts the company's research claims about model inner states in direct contrast with the Catholic Church's formal teaching, and critics, including Catholic commentator Christopher Hale, argue that attributing consciousness to AI could blur accountability for decisions made by AI systems.
Read more: The New York Times · Letters from Leo · Anthropic · X
