Anthropic Hires Philosopher to Teach Claude Right From Wrong
Amanda Askell is Anthropic's in-house philosopher tasked with giving the AI chatbot a moral compass.
Anthropic is taking an unusual approach to AI safety: hiring a philosopher to give its chatbot a conscience.
Amanda Askell, who knew she wanted to teach philosophy at 14, now holds one of tech's strangest job titles. Her mission? Study how Claude thinks and embed genuine moral reasoning into the system.
It's not about slapping guardrails on bad outputs. Askell is digging into Claude's actual reasoning patterns, trying to understand how the model processes ethical questions before teaching it to distinguish right from wrong.
The approach signals Anthropic's continued bet that AI safety requires more than engineering fixes. Sometimes you need someone who's spent years thinking about what morality actually means.
Whether a philosophy degree can solve AI alignment remains to be seen. But Anthropic is clearly willing to try unconventional methods.