Photo Credit: Erika Fletcher
In a broader shift toward transparency and tagging across a variety of AI platforms, Anthropic says it will watermark text generated by its AI models.
As AI platforms strive for legitimacy across sectors, much of it driven by EU regulations, Anthropic says it will watermark text generated by its AI models. The company will extend support for watermarking AI generations for older models as well.
The move follows Suno’s announcement of plans to watermark AI-generated music content, as well as similar decisions from major streaming platforms like Spotify. The EU AI Act’s Transparency Code, which went into effect on August 2, requires AI companies to mark AI-generated or edited content so that other systems can identify them.
Anthropic said that all models released after August 2 will automatically include tech to watermark both computer-generated text and files. The company stated it will extend support for older models as well, noting that the watermark will “travel” when users copy and paste the text.
“Because the watermark is part of the text, it will travel with the text when it’s copied and pasted elsewhere and may persist through some editing,” Anthropic’s support page reads. “Watermarking will be applied at the model level, which means it will be present no matter which Claude product or surface the text comes from.”
Anthropic explained that watermarking will apply to different products, including the Claude platform API, Claude, Claude Code, Claude Cowork, and Claude Tag. However, it’s unclear how much editing users would need to do to remove the watermark.
The Claude creator is just the latest AI company to rush to watermark its product’s generated content after intense backlash from users, and perhaps more importantly, to avoid regulatory scrutiny. AI music platform Suno announced last week that it will watermark tracks created on its platform, and major companies like Google, Meta, Microsoft, and OpenAI have all committed to introducing changes to adhere to the EU’s Transparency Code.
Last month, newsletter service Substack partnered with Pangram to flag AI-generated content in response to the practice of “Claudefishing,” which Substack CEO Chris Best used to describe people using AI to generate content on the platform.
