In a recent LinkedIn post, Federico Donatone discusses Anthropic’s new initiative to invisibly watermark AI-generated text, expressing significant concerns about the implications for users and content creators.
Donatone highlights the announcement that Claude will embed invisible watermarks within its AI-generated text, a move that Anthropic states is designed for transparency and detection. These watermarks are reportedly capable of surviving copy-pasting and some editing, with similar metadata being applied to AI-generated images and files.
“Claude will embed invisible watermarks in AI-generated text.”
Concerns Over User Control and Transparency
Federico Donatone raises immediate objections to this new policy, particularly from the perspective of a paying customer. He argues that the embedding of invisible watermarks represents an unwanted form of control over user-generated content.
As Donatone pointed out:
“If I’m paying for your plan, I don’t want invisible watermarks embedded in my content.”
This sentiment underscores a potential conflict between Anthropic’s stated goals of transparency and the user’s expectation of unfettered use of a service they are paying for. Donatone suggests that such impositions could drive users towards alternative solutions.
The Push Towards Open Source Alternatives
In light of Anthropic’s direction with Claude, Federico Donatone suggests that the development could strengthen the appeal of open-source AI models. He posits that if invisible watermarking becomes a standard feature, users seeking greater control over their output might actively seek out alternatives that do not impose such restrictions.
Donatone’s view is clear:
“If this is really the direction they choose, we should seriously consider open source alternatives that don’t impose this kind of control.”
This perspective positions the decision by Anthropic not just as a technical implementation but as a strategic choice that could influence market dynamics within the AI landscape.
Critique of Anthropic’s Decision-Making
Federico Donatone does not hold back in his critique of Anthropic’s decision, framing it as a continuation of what he perceives as problematic choices by the company. He expresses a sentiment of surprise, suggesting that he has previously underestimated the company’s capacity for making decisions he views negatively.
In Donatone’s words:
“I keep thinking Anthropic can’t possibly top their last bad decision but here we are.”
This strong statement reflects a deep-seated skepticism about Anthropic’s strategic direction and its impact on the user experience and broader AI ecosystem. Donatone’s analysis prompts a discussion about the balance between AI safety, transparency, and user autonomy in the rapidly evolving field of artificial intelligence.
📝 About This Content
This article is based on insights shared by Federico Donatone on LinkedIn.
📅 Originally posted on August 12, 2026 | View original post on LinkedIn →