Anthropic, OpenAI turn to faith in search for ethics

Anthropic and OpenAI are turning to faith as they strive to create ethical artificial intelligence. These companies are recognizing that religion can provide guidance on how to make chatbots behave morally. However, experts warn that this approach may not fully capture the complexities of religion and its impact on different cultures and worldviews.

Companies like Anthropic, OpenAI, Meta, Google, and Amazon are seeking input from philosophers and faith leaders to inform their AI models. OpenAI CEO Sam Altman has stated that the goal is to create a moral framework that reflects an aggregate of humanity's moral views. However, scholars and adherents alike argue that this approach fails to account for the diverse nature of religious traditions and the way they intersect with culture and politics.

Anthropic, for example, has met with scholars, clergy, philosophers, and ethicists from over 15 religious and cross-cultural groups. These meetings aim to broaden representation and encourage diverse perspectives. Meanwhile, other tech giants have also engaged with the Vatican and religious leaders to address ethical concerns surrounding AI.

While these efforts may appear thoughtful, experts caution that reducing religion to a set of moral principles risks overlooking the richness and complexity of religious traditions. Religion is not merely a rulebook, but a deeply personal and cultural experience that cannot be easily distilled into a formula for AI training.