Constitutional AI

13 posts · 1 source

AnthropicBlog13 Jul

How Claude's values vary by model and language

AnthropicBlog8 May

Teaching Claude why

AnthropicBlog30 Apr

How people ask Claude for personal guidance

AnthropicBlog23 Feb

The persona selection model

AnthropicBlog19 Jan

The assistant axis

AnthropicBlog9 Jan

Next-generation Constitutional Classifiers: More efficient protection against universal jailbreaks

AnthropicBlog21 Apr

Values in the wild: Discovering and analyzing values in real-world language model interactions

AnthropicBlog3 Feb

Constitutional Classifiers: Defending against universal jailbreaks

AnthropicBlog18 Dec

Alignment faking in large language models

AnthropicBlog17 Jun

Sycophancy to subterfuge: Investigating reward tampering in language models

AnthropicBlog24 Oct

Specific versus General Principles for Constitutional AI

AnthropicBlog9 May

Collective Constitutional AI: Aligning a Language Model with Public Input

AnthropicBlog15 Dec

Constitutional AI: Harmlessness from AI Feedback