The Hindu monk teaching Claude right from wrong

🇮🇳 The Times of India (IN) —
The Hindu monk teaching Claude right from wrong

AI Summary

Anthropic uses a set of broad principles, described as a “constitution,” to guide Claude’s behavior in situations its creators may not have anticipated. A Hindu monk is involved in teaching the AI system right from wrong, though the description provides no further details about his role.

Anthropic does not want Claude merely to obey an enormous list of prohibitions. It already uses a “constitution”, a set of broad principles designed to guide the model through situations its creators cannot predict in advance.

AI & Tech Anthropic Claude AI ethics AI alignment constitutional AI

Read original source →