OpenAI’s Altman draws regulatory divide with Anthropic over AI risks: Politico

OpenAI CEO Sam Altman has voiced his discomfort with anthropic efforts to imbue artificial intelligence with religious or spiritual characteristics. This stance appears to be a direct response to Anthropic's research exploring the possibility of AI consciousness and the potential impact of religious ideas on AI behavior. Altman raised safety concerns, stating that he is "very uncomfortable" with the notion of ascribing religious force or surrendering human judgment to AI models.

His comments were widely perceived as a criticism of Anthropic and its co-founder Dario Amodei, who previously collaborated with Altman at OpenAI.

The disagreement extends beyond the question of whether AI can think or feel. It delves into the possibility that future AI systems could possess consciousness, emotions, or a form of moral status, prompting the question of whether humans should treat them differently if they do. Anthropic's exploration of these concepts intensified after a New York Times report revealed that Anthropic co-founder Christopher Olah had been discussing AI consciousness with religious leaders and contemplating the integration of religiously-inspired moral behavior into AI models.

The debate has gained momentum within the technology industry, with some prominent AI researchers openly questioning the moral and spiritual standing of AI models. Google DeepMind principal research scientist Jon Barron criticized this approach in a post, expressing concern that researchers are elevating the moral status of AI components without focusing on solving real-world problems and assisting humans.

The discussion has also caught the attention of Pope Leo, who has cautioned against the potential risks AI poses to human creativity, urging the Catholic Church to collaborate with artists to safeguard it.

This emerging debate arises amid traditional AI safety discussions that primarily focused on more immediate risks, such as cyberattacks and the use of AI to aid in the development of biological weapons. However, this latest controversy raises a more profound question about the future of AI: If machines were to become conscious, should humans treat them as more than mere tools, and could religion play a role in determining how they should behave? OpenAI and Anthropic have not yet responded to requests for comment on Altman's post.