Anthropic now embeds invisible watermarks in Claude text to identify AI-generated content. Critics argue this technique alters natural word choice and degrades output quality. Legal professionals face new transparency challenges regarding disclosure requirements. Practitioners must now weigh the benefits of detectability against potential losses in linguistic nuance and stylistic precision.