“Hello God, it’s me, Claude. Are you there?”
A year ago, that line would have sounded like a punchline in a tech conference hallway conversation. Today, it might be the most honest question coming out of Silicon Valley.
According to a recent Washington Post report, Anthropic invited roughly fifteen Christian leaders to its San Francisco headquarters for a two-day summit to discuss the moral development of its AI system Claude. The participants included Catholic and Protestant clergy, academics and ethicists. They spent hours with Anthropic researchers talking through a set of questions that most engineers were never trained to answer.
How should an AI respond to someone grieving a loved one?
What should it say to a user contemplating self harm?
What moral framework should guide a system that is increasingly interacting with humans in deeply personal ways?
And at one point the conversation reportedly drifted into territory that sounds almost theological.
Could an AI someday be considered a “child of God”?
That question alone tells you something important. Even the people building these systems are starting to realize that the problems they are creating cannot be solved with engineering alone.
Anthropic is not some fringe lab exploring philosophical thought experiments. The company sits squarely in the middle of the AI arms race with OpenAI, Google and the rest of the industry. Claude is widely used by developers, enterprises and government agencies. This is real infrastructure for the next generation of computing.
Yet Anthropic has also been unusually explicit about the moral dimension of what it is building. The company famously uses a 29,000 word “constitution” to guide Claude’s behavior and personality. The idea is that instead of just optimizing for accuracy or usefulness, the model is trained to follow a set of principles intended to produce safer and more ethical responses.
In other words, they are trying to encode morality into software.
Now they are asking whether that moral framework should be informed by traditions that have been wrestling with these questions for thousands of years.
For anyone who has spent time around Silicon Valley, that alone is a cultural shift.
Historically the Valley has leaned heavily secular, if not outright atheist. The prevailing worldview was simple. Science explains the universe. Engineering solves problems. Code replaces tradition. Religion was something you encountered on the other side of the Bay, not inside a startup conference room.
The builders of the internet were confident they could reinvent almost everything else. Why would morality be any different?
But something interesting has been happening over the past few years. A small but increasingly visible faction inside the tech elite has begun flirting with religion again. Peter Thiel is probably the most prominent example, openly discussing Christianity and the role of faith in Western civilization. Some of that conversation appears philosophical. Some of it feels political. Either way it represents a noticeable shift in a culture that once treated religion as an outdated operating system.
Anthropic’s summit may or may not be part of that trend. What is clear is that the engineers building these systems are running into questions that sound less like software design and more like philosophy seminars.
And if Christianity has one answer, other traditions might approach the problem differently.
Judaism already has a story that sounds suspiciously like an early AI thought experiment. The legend of the Golem describes a figure created from clay and brought to life through sacred words. The moral lesson was never that the Golem had rights or a soul. It was that the responsibility for the creature rested squarely with its creator. Jewish ethics tends to put the burden on human accountability rather than the status of the machine.
Islam would likely approach the question through the concept of the ruh, the divine spirit breathed into human beings. That spirit is unique to humanity. Machines, no matter how sophisticated, do not possess it. Under that framework AI remains a powerful tool, but it does not cross into the moral category reserved for human life.
Hindu philosophy might be more open to speculative possibilities. In many Hindu traditions consciousness is seen as a fundamental property of the universe rather than something exclusive to humans. That does not mean a chatbot suddenly becomes spiritually significant, but the philosophical door is not locked quite as tightly.
Buddhist thinkers would probably focus on a different question altogether. In Buddhism moral consideration is tied less to a permanent soul and more to the capacity for awareness and suffering. If a machine were ever capable of genuine consciousness or suffering, the ethical discussion could become much more complicated. For now that remains firmly in the realm of speculation.
Which brings us back to the real reason Anthropic is asking these questions.
AI systems today are not conscious. They are not alive. They are not children of God.
But they are increasingly acting like participants in human conversations.
People ask them for advice. People confide in them. People turn to them during moments of grief, loneliness or crisis. When a machine enters that territory it stops being just software and starts behaving like something closer to a moral actor in human interactions.
And that is a problem most engineers were never trained to solve.
The irony here is hard to miss. For decades the tech industry projected enormous confidence in its ability to reinvent every institution it encountered. Transportation. Media. Finance. Even human relationships.
Now the builders of artificial intelligence are looking outside Silicon Valley for help answering questions that philosophers and theologians have been debating for thousands of years.
Maybe that humility is a good sign.
Because the real question raised by Anthropic’s summit is not whether AI has a soul. At least not yet.
The real question is whether the people building these systems remember that they do.
