Anthropic's Claude Gets a Moral Compass: Philosopher Amanda Askell Shapes AI Ethics

• Anthropic entrusts philosopher Amanda Askell with shaping the moral compass of its chatbot, Claude.
• Askell crafts extensive prompts to guide Claude's behavior across millions of weekly conversations, focusing on character and ethics.
• Her work aims to ensure Claude distinguishes right from wrong, reads social cues, and resists manipulation.
• Anthropic's unique approach prioritizes AI character, contrasting with industry trends of dispersed safety teams.
• This focus addresses growing concerns about AI's unintended consequences, including misuse and unhealthy relationships.