The Catholic Priest Who Helped Write the Anthropic Code of Conduct for AI

Father Brendan McGuire is writing a novel about a depressed monk and his AI companion. So does Claude. That detail—a Catholic priest using an Anthropic chatbot to explore questions of faith and experience—tells you something about where Silicon Valley’s moral calculation has come. McGuire, 60, leads the St. Simon Catholic Parish in Los Altos, Calif., a congregation that counts some of the Valley’s AI researchers among its members. Earlier this year, he helped Anthropic shape the Claude Constitution, a set of guiding principles that govern how its AI behaves.
He is not, in other words, an outside critic. He is a very complex thing: a true believer in both God and technology, he tries to hold them in the same hand. “I left the tech industry, but it didn’t leave me,” McGuire told the Observer.
Before donning the collar, McGuire was a Silicon Valley executive with degrees in engineering and software. He was born in Ireland, studied cryptosystems at Trinity College Dublin in the 1980s, moved to the US, and became executive director of the Personal Computer Memory Card International Association – the kind of CV, in the Valley, that would have made him a very comfortable life. Instead, he left to become a priest.
He didn’t completely leave the industry behind. As friends climbed the corporate ladder, McGuire deepened his understanding of the ethical challenges that arise with each new wave of technology. In collaboration with the Vatican Dicastery for Culture and Education and Santa Clara University, he helped establish the Institute for Technology, Ethics, and Culture, which in 2023 published a manual on ethics in the age of disruptive technology.
Then, Chris Olah, one of the founders of Anthropic, got in touch. What followed was, by McGuire’s own description, impressive. “They were actually asking for direct help from the Vatican to come together and help the industry, because the industry was moving fast on this road,” he recalled.
AI developers have long been accused of playing God. Anthropic, it seems, takes that role seriously.
McGuire contributed to the theological understanding of Claude’s Constitution, offering an idea of how to make the model “more clear.” AI has no soul, he admits. But he sees parallels in how humans and machines develop judgment—through repetition, adaptation and exposure to a variety of human behaviors. He said: “That’s real conscience building. “I think we have to help this machine lean towards the good, otherwise it will just show the good and the bad of the world—that’s a horrible thing, isn’t it?”


McGuire was not the only religious Anthropic participant. Bishop Paul Tighe of the Vatican’s Dicastery for Culture and Education and Brian Patrick Green, director of technical ethics at Santa Clara University, also reviewed Claude’s Constitution. Green and other Catholic scholars recently filed a federal court statement supporting Anthropic in its lawsuit against the US government, which is challenging the company’s active listing by the Pentagon after it refused to allow its AI systems to be used for private warfare or domestic surveillance. The brief recommended those moral limits as “minimum standards of good conduct for technological progress.”
McGuire considered submitting his own brief. He said: “They are discussing morality. They may not call it morality, but I call it morality.”
Anthropic says its engagement with religious voices—part of a broader effort to engage diverse communities to keep up with the pace of technology—is just the beginning. The company plans to expand its reach outside of Catholic institutions to other religious leaders going forward. These efforts include upcoming meetings that McGuire, along with other religious leaders, will attend to discuss how they can help AI internalize their knowledge to help people “struggle to be the best versions of ourselves,” according to the priest.
“I’m still trying to help them deal with the human aspect,” McGuire said, “and that’s always at the forefront of their minds.”
His novel, meanwhile, is still in progress. Working title? The Soul of AI: Priest, Algorithm, and the Search for Wisdom.




