Claude 1

Quick Facts

• Released: 2023
• Focus: Safety and alignment
• Capabilities: Strong reasoning, cautious outputs
• Status: Archived (replaced by Claude 3)

Key Achievement

First production model using Constitutional AI approach for alignment.

Performance

• Good general reasoning
• Excellent safety properties
• Conservative in harmful requests
• Strong on ethical reasoning

Current Status

Superseded by Claude 2 and 3, but historically important for proving Constitutional AI works.