Anthropic has begun watermarking outputs from its Claude chatbot, embedding invisible code into text to mark it as AI-generated. This move, reported by TechCrunch, responds to the EU AI Act’s Transparency Code, which requires companies to label AI-generated or edited content in ways detectable by computer systems. European regulators see this as a step toward accountability in AI use across Europe. However, some users have voiced frustration on Reddit, arguing the watermarking unfairly exposes everyday use cases, with critics calling it a “digital tattoo on the forehead” and “unethical” and “disgusting.” Others defended the practice as a sensible safeguard, with one commenter stating, “There is literally no good argument for why this isn’t a good idea.” Separately, Anthropic announced that auto mode will become the default setting for its Claude Code platform starting August 14, affecting Pro, Max and Team accounts. Developers will no longer need to manually approve each action, unless it is deemed ‘irreversible, destructive, or aimed outside your environment’. First introduced in March as a test feature, testing showed auto mode caught 89% of harmful actions compared to just 13.6% detected by human reviewers. The company noted users tend to approve nearly all permission prompts manually, making automated oversight more effective.