Raised in Canada, Olah worked on projects related to 3D printing after receiving a Thiel Fellowship in 2012. The program, founded by conservative billionaire Peter Thiel, pays young people to work on a start-up or research project after skipping or stopping college. After finishing his two years as a Thiel fellow, Olah went on to an internship at Google, where he helped develop DeepDreams, a neural network project that created psychedelic art.

At OpenAI, Olah ran an interpretability research lab, where he designed projects that helped explain what was going on as the team built its large-language model. In late 2020, he followed Amodei out of the company. On his X account, Olah frequently reposts comments from other employees who have left OpenAI, purportedly because they’re not satisfied with the company’s commitment to safety. When Anthropic was engaged in a public battle with the Department of Defense over the use of its models, Olah also reshared an amicus brief filed by Catholic moral theologians in support of Anthropic’s case.

Olah is an active blogger and seems to be a part of the social network that has sprung up around the pursuit of artificial intelligence in the Bay Area. In 2022, he wrote a blog post about his search for a romantic partner, which was picked up by Wired. In the post, he mentions his passion for the “beauty of math and science,” his search for “wholesomeness” in his life, and his “politically moderate” leanings. Olah also said that he values “moral connection” and discussed his veganism. He mentioned his mixed emotions about the Effective Altruism movement, in which tech-inclined people organize for data-driven charitable donations.

“I’m very dedicated to doing what I believe to be morally right. I think some people find me frustratingly scrupulous, but I think for others I can help them feel supported in holding true to their own values (even when we differ)” Olah noted in the blog post. “However, there are also some things I’d disagree with the EA mainstream on, and I generally value moral independence.”

The post may help to explain why Olah was Anthropic’s pick for the pope. In it, Olah says he was raised as an evangelical Christian with creationist leanings, until he became an atheist around the age of 15. In his remarks at the Vatican, he demonstrated some facility with the language of the Christian tradition, citing concepts like discernment, human flourishing, and the church’s history of concern for the global poor. But his remarks were secular in nature—and in them, he also conceded that his company’s tools present a challenge to the future of human labor.

“We dwell so often on what divides us, but humanity, full of dignity and conscience, has so much common ground,” Olah said before Pope Leo. “In conversations we at Anthropic have had with leaders across faith and cultural traditions, we found one shared and deeply held conviction: if this technology is coming, it must go well—for our common home, and for the children to come.”