In Depth
Developed by Anthropic, constitutional AI reduces the need for human feedback at every step by giving the model a set of rules it can self-evaluate against. The model critiques its own outputs and revises them before responding.
Top Stories
All Sections
Resources & Tools
About
Developed by Anthropic, constitutional AI reduces the need for human feedback at every step by giving the model a set of rules it can self-evaluate against. The model critiques its own outputs and revises them before responding.