At a Glance · AI-generated

Anthropic announced on August 10 that it will embed invisible watermarks into all text and images generated by Claude, its AI assistant, making AI-generated content detectable even after editing. The company released a watermark detection API allowing third parties to identify Claude outputs, positioning the move as part of industry efforts to combat AI-generated misinformation. The rollout sparked mixed reactions: some users praised the transparency measure, while others canceled subscriptions citing concerns about detection of AI use in professional and academic contexts. Anthropic defended the feature as balancing detectability with output quality, distinguishing its approach from competitors still developing watermarking tools.