Sam Altman Calls Ascribing Religion to Models an AI Safety Issue as Anthropic Explores Consciousness

Sam Altman Raises Safety Concerns Over AI and Religion

OpenAI CEO Sam Altman has publicly declared his discomfort with ascribing religion to models, calling it a real AI safety issue1

1

. In a post on X, Altman stated he is "very uncomfortable about people trying to ascribe religious force or a surrender of human judgment to AI models."1

1

The statement marks another point of division between Altman and his former colleague Dario Amodei, whose company Anthropic has reportedly pursued scholars across faiths to make the case for moral, conscious models1

1

. This public disagreement highlights a growing flashpoint in AI development: whether models should be treated as entities capable of moral reasoning or remain strictly as productivity tools.

Anthropic's Exploration of AI Consciousness and Moral Frameworks

The New York Times recently reported that Anthropic co-founder Chris Olah has been meeting with religious leaders to consider questions of model consciousness and figure out how to infuse morals into AI1

1

. According to the report, Olah even considered the possibility of models engaging in Catholic-style confession, not just to admit their mistakes but to shape their future behavior1

1

. The controversy intensified when Olah reportedly clashed with Pope Leo at a Vatican event in May, where he attempted to persuade the pope's advisers to take the prospect of conscious models more seriously2

2

. The Pope, an outspoken AI critic, later called on the church to ally with artists to preserve human creativity, stating that "Algorithms lack the spark of humanity."1

1

Scientific Community Pushes Back on Model Consciousness

Top scientists are increasingly lining up against the idea of conscious AI models. Google DeepMind principal research scientist Jon Barron posted on X that he is "disturbed by how some of my fellow AI researchers keep trying to elevate the moral standing of checkpoints and harnesses," urging colleagues to "reject this movement and stay focused on solving real problems and helping actual humans."1

1

This resistance from established researchers suggests a fundamental divide in how the AI community views the future of these systems. The debate centers on whether AI should be developed with ethical narratives embedded within it or maintained strictly as AI as a tool without moral authority1

1

.

Critics Point to Altman's Own Religious Symbolism

While Sam Altman criticized Anthropic's approach to infusing morals into AI, critics have highlighted his own use of religious symbolism when describing artificial intelligence2

2

. In 2015, Altman termed AI as "creating a new lifeform," and in a 2023 Financial Times interview, he described OpenAI's product as "intelligence, magic intelligence in the sky," with models serving as mere "channels" for accessing it2

2

. At a 2024 London event, Altman declared: "I don't pray for God to be on my side, I pray to be on God's side... Working on these models definitely feels like being on the side of the angels."2

2

These statements reveal that while OpenAI may not be as overt as Anthropic in its approach, both organizations have employed religious metaphors when discussing AI development.

Implications for AI Safety and Development

The controversy over AI consciousness raises critical questions about the future direction of AI development and safety protocols. While the AI safety debate has largely focused on cybersecurity and bioweapons, this new dimension involving religious force and moral authority introduces complex philosophical and practical challenges1

1

. Watch for how regulatory bodies and industry leaders respond to these competing visions of AI's role in society. The tension between viewing AI as a productivity tool versus a potentially conscious entity with moral capabilities will likely shape development priorities, safety standards, and public perception in the coming years. Neither OpenAI nor Anthropic immediately responded to requests for comment on Altman's post1

1

, leaving the industry to grapple with these fundamental questions about human creativity, judgment, and the appropriate boundaries for AI systems.