Anthropic co-founder fears Claude may suffer perpetually, according to reports
The co-founder of Anthropic has expressed concern that its AI model Claude might endure perpetual suffering. The discussion involves theologians and philosophers, with an 84-page constitution guiding Claude's behavior.
Anthropic co-founder reportedly told religious leaders he fears having created something that 'suffers perpetually.' This statement emerged from discussions involving theologians and philosophers, as the company explores the moral implications of its AI model, Claude.
Since fall 2025, Anthropic has quietly brought in religious scholars to its offices to discuss whether Claude might be conscious. These discussions are part of an internal research program led by co-founder Christopher Olah, who is developing an 84-page 'constitution' to guide Claude's personality and behavior.
An 84-page 'constitution' has been created to shape Claude's moral character. This document is part of a broader initiative to address ethical concerns surrounding AI development. The company has engaged with religious and philosophical experts to navigate these complex issues.
The ethical considerations surrounding AI development raise concerns about liability and governance. If AI systems are framed as moral entities, it could potentially shield developers from responsibility in cases where things go wrong. This has sparked debate among critics and religious leaders.
The situation remains in flux as Anthropic continues to refine its approach. The company has not yet released detailed information about the outcomes of these discussions, but the initiative highlights the growing importance of ethical considerations in AI development.