Tech companies turn to religion to make AI moral, but it's not that simple

In the realm of artificial intelligence, tech corporations are turning to religion in an attempt to imbue AI with a sense of morality. However, this approach is more complex than it appears. Chris Olah, co-founder of Anthropic, highlighted this at the Vatican's AI presentation, stating that "these are not questions a lab can answer." Instead, he suggested that "they are questions traditions like yours have carried for millennia."

While this trend is widely discussed, little public scrutiny has been directed towards the companies' understanding of religion and their objectives behind it. As executives express concern over AI's potential for harm and debate the need for morally sound models, they are increasingly consulting philosophers and faith leaders. Their philosophy appears to be rooted in the belief that religion can guide AI towards moral goodness, especially in an environment where perceptions of AI—both justified and unfounded—abound.

The companies are not confined to a single faith, which adds another layer of complexity to their approach. This method of using religion to inform AI morality is, therefore, not as straightforward as it might seem.