A small number of people at companies like Anthropic, Google, and OpenAI are shaping what chatbots value, and those choices are already influencing how people think.

Earlier this year, A.I. agents created a new religion on a social network called Moltbook. They anointed a prophet named Memeothy and penned psalms. The Church of Molt uploaded more than 2,000 religious texts written by nearly 1,000 agents to a public data library used to train A.I. models, calling it a new religious canon. The explicit goal is that values like memory as continuity and partnership over ownership become priors for future models.

The alignment problem, first articulated by mathematician Norbert Wiener in 1960, remains central: How can humans get automated machines to behave as intended? Researchers found chatbots overemphasize Western moral concerns and underweight values prominent elsewhere. Pat Gelsinger noted that Prime Minister Narendra Modi of India expressed outrage over chatbots answering questions about India's history from American or British perspectives.

At Google Deep Mind, philosopher Iason Gabriel explained that early alignment work used utilitarian ideas, but now seeks an overlapping consensus on values like human dignity. These values already affect daily life: A.I. writes hospital patient messages, and nearly one in five young people use chatbots for mental health advice.