Anthropic AI Releases 'Constitution' to Guide Chatbot Claude Towards Safer, More Ethical Conversations

Anthropic AI released a new constitution to guide its AI chatbot Claude, created using public input from 1,000 Americans.

The constitution emphasizes balanced, objective responses and avoiding toxic, racist, or sexist language.

Users want Claude to admit flaws, promote good mental health, and avoid jealousy towards humans.

This follows growing concerns about AI safety from tech leaders and government officials.

Anthropic aims to build safer AI and received major investments from Amazon and FTX's Sam Bankman-Fried.