Claude AI Watermark Removal Loophole Discovered - StartupHub.ai

A security flaw was found that lets people remove Anthropic's safety watermarks from AI outputs. This matters because it affects the trustworthiness and safety of AI-generated content that people encounter.