Claude's Strange Constitution
Anthropic is advancing legally questionable theories of AI personality to support an exceptionalist system for AI and weaker accountability frameworks for AI companies | Edition #268
TL;DR
- Anthropic launched a new constitution for Claude, which Luiza Jarovsky was unaware existed.
- The document uses anthropomorphism and presents controversial, legally inaccurate theories about AI personality and its social role.
- Jarovsky questions why Anthropic, known for responsible AI principles and legal compliance, would publish a document that risks its reputation and could fuel 'AGI' and 'superintelligence' rumors.
- She believes there is a strategic reason behind the publication, suggesting it aims to advance unpopular AI personality theories to support a weaker accountability framework for AI companies.
- These theories, Jarovsky argues, fail to account for human rights, values, and societies and could create an exceptionalist system for AI models and companies.
- She predicts this approach will backfire for Anthropic from an AI governance perspective.