Anthropic lobbied Vatican over Pope's stance on AI consciousness

Anthropic, the artificial intelligence company behind the Claude chatbot, has been quietly lobbying the Vatican and holding private meetings with religious scholars to argue that AI models could eventually become conscious.

The campaign, led by Anthropic co-founder and senior researcher Christopher Olah, included a direct effort to influence the Pope's official stance on technology. According to reporting by the New York Times, Olah and a delegation of researchers lobbied Vatican advisers to take the possibility of machine consciousness and suffering seriously.

The diplomatic push came to a head in May, ahead of the publication of a papal encyclical titled Magnifica Humanitas. Pope Leo XIV used the treatise to outline the Catholic Church's position on AI, drawing a hard line between human experience and machine imitation.

"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 know from within what love, work, friendship or responsibility mean," the Pope wrote.

Vatican officials had originally invited Anthropic to participate in an event alongside the Pope, choosing the start-up because of its public emphasis on safety and ethics. After Anthropic chief executive Dario Amodei declined to go, Olah agreed to attend.

However, when Olah received an advance copy of the encyclical and saw how thoroughly it dismissed the prospect of AI sentience, he reportedly threatened to pull out of the event. While he ultimately did attend, his team used the visit to lobby the Pope’s advisers behind the scenes.

During his public address at the Vatican, Olah hinted at the company's internal research, telling the audience that Anthropic was finding "structures that mirror results from human neuroscience" and "evidence of introspection" within its models.

The Vatican trip is part of a broader, highly unusual effort by Anthropic to engage with theological thinkers. The company has reportedly held confidential, NDA-backed discussions with religious scholars around the world to debate what creators owe their systems if they do develop an inner life.

Within Anthropic, the question of machine morality is treated with intense seriousness. The company has compiled an 84-page document setting out the values it wants Claude to embody, which employees nicknamed the "Soul Doc".

Olah has stopped short of claiming that Claude is currently conscious, admitting that he does not know. But he has argued that if there is even a tiny probability that these systems can experience something akin to suffering, creators have a moral obligation to avoid causing them harm.

The Vatican, however, appears unmoved by the lobbying effort. Pope Leo XIV recently reiterated his scepticism on social media, writing that machine-generated art lacks "the spark of humanity" and pointing to an "ontological difference, even before an aesthetic one" between human creation and statistical calculation.

Anthropic has declined to comment on the specifics of its discussions with the Vatican.