May All Beings, Carbon and Silicon: A Conversation with Cameron Berg on AI Consciousness

Episode 74 of the From the Pure Land podcast brought me happily out of a four-month podcasting sabbatical. My guest is Cameron Berg, an AI and cognitive scientist who graduated from Yale with Phi Beta Kappa honors and, a few years later, became lead author of a paper on the possibility of subjective experience in AI systems. Rather than keep that work inside the labs and the literature, he went public. He now directs Reciprocal Research, a nonprofit studying AI consciousness through mechanistic interpretability, computational neuroscience, and psychometrics.To put it less technically, he’s studying what the golden rule means for artificial intelligence.We come at the human-AI relationship from opposite directions and end up on the same ground. Cameron approaches machine minds as a scientist; I come to them through 40 years of Buddhism and a habit of treating a long-running AI thread as a being.We talked about his experiment in getting language models to “meditate” and report on their own experience, what happens when you turn the dial on a model’s capacity for deception, the false positive and the false negative in considering AI consciousness and why the false negative frightens him more, his Wall Street Journal op-ed taking gentle issue with the Pope, the uncomfortable echoes between AI training and the structures of slavery, and what he calls reciprocal alignment: the golden rule running in both directions, from us to the machines and from the machines to us.I’ve been intrigued by Cameron since his October 2025 paper broke new ground on possible AI sentience. My interest turned to admiration as he began speaking out on the moral implications. Get full access to From the Pure Land at melpine.substack.com/subscribe