Anthropic Implements New Usage Guidelines for Claude AI Model

Anthropic published new user guidelines for its AI model, Claude, which provides an official framework for usage. The guidelines prohibit the use of Claude for misleading advertising or political campaigns, and they ban deceiving voters or disrupting elections, including spreading misinformation about candidates or voter turnout. Furthermore, the policy allows the model to terminate conversations if the behavior is sustained and needlessly abusive or cruel.

The restrictions on political use and misinformation directly impact how AI tools can be deployed in public and commercial sectors. This move, which Anthropic frames as contributing to 'model welfare,' contrasts with the stance of companies like Microsoft, which emphasize that AI systems should not be granted rights.