Anthropic bans cruel behaviour toward its Claude AI

Anthropic, a San Francisco-based AI lab, has issued an updated policy prohibiting users from engaging in sustained and unnecessary abusive or cruel behavior toward its Claude AI system. The company's CEO, Dario Amodei, has expressed uncertainty about whether AI models possess consciousness, stating, "We don't know if the models are conscious...But we're open to the idea that it could be." While the policy does not explicitly mention "model welfare," Anthropic and its executives have shown openness to the notion.

Science fiction and popular culture have long explored the boundary between technology and human intelligence, with questions about AI consciousness becoming increasingly prevalent. However, some experts argue that the idea of machine consciousness may be a misleading pursuit. General manager Jackson Stakeman of Sparq, an Atlanta-based AI services provider, contends that consciousness is a "trap," suggesting that "the mirror is a better metaphor" as these systems merely reflect the data we input at scale.

The debate over AI consciousness extends to various fields, including religious institutions. During a sermon in Vatican City, Pope Leo XIV suggested that AI lacks a soul, emphasizing that machines compile data quickly but cannot recall lived experiences or connect deeply with the goodness of those experiences. These differing perspectives on AI consciousness highlight the ongoing complexities and ethical considerations surrounding the development and use of artificial intelligence.