The rapid advancement of generative artificial intelligence has prompted leaders in Silicon Valley to reassess their traditionally skeptical views toward religious institutions. Developers of neural networks are facing a significant challenge: government regulations are lagging behind technological innovations, making the establishment of ethical guidelines for AI systems increasingly urgent.
A historic roundtable, named the Faith-AI Covenant, took place in New York, gathering executives from major AI firms such as Anthropic and OpenAI alongside leaders from various global faiths. The meeting was organized by an interfaith alliance based in Geneva that specializes in combating extremism and human trafficking.
The primary goal of this unexpected collaboration is to develop a unified set of human principles that future AI agents will adhere to. Following the initial discussions in the United States, similar meetings are planned for Beijing, Nairobi, and Abu Dhabi.
Baroness Joanna Shields, a key figure behind the initiative and a former executive at Google and Facebook, emphasized that technology developers are acutely aware of the power and risks associated with their creations. She believes that world religions, with their billions of followers, possess unique expertise in shaping ethical frameworks.
Some religious organizations have proactively engaged with AI issues even before being officially invited by tech giants. For example, the Church of Jesus Christ of Latter-day Saints has included AI in its official handbook, stating that while technology cannot replace divine inspiration, it can serve as a valuable educational tool. The Southern Baptist Convention has also adopted a resolution calling for proactive measures to establish ethical frameworks for new technologies rather than merely reacting to their impacts on communities.
Among the tech companies, Anthropic has taken the most active role in involving religious leaders. The creators of the popular chatbot Claude have publicly stated that their internal ‘AI Constitution’ was developed with direct input from ethics and theology experts. This move is also seen as a crucial step for Anthropic to restore its reputation following a public dispute with the Pentagon over the prohibition of using their models for creating autonomous weapons.
Despite the good intentions of the participants in this initiative, it faces significant criticism from both religious scholars and independent AI researchers. A primary concern is that, even with shared foundational values, the priorities of different religious communities vary considerably. Creating a universal moral code for regulating advanced technologies that satisfies Christians, Sikhs, Jews, and Buddhists alike is viewed as virtually impossible.
Furthermore, some in the scientific community regard this engagement with religion as mere public relations. Experts recall the old Silicon Valley adage, ‘Move fast and break things,’ noting that companies have disrupted numerous social institutions and are now seeking justification for their actions.
Former U.S. administration representative Rumman Chowdhury describes the religious roundtables as an attempt to evade real legal issues and regulatory challenges, suggesting that they shift the responsibility for ambiguous ethical dilemmas in AI development onto spiritual leaders.
Tech leaders and religious institutions are collaborating to establish ethical guidelines for AI, addressing the urgent need for moral frameworks in technology. However, the initiative faces criticism over the feasibility of creating universal principles applicable across diverse religious beliefs.