War on watermarks: users mass-cancel Claude subscriptions due to forced text labeling

Implementation of mandatory content labeling

On August 16, 2026, the artificial intelligence laboratory Anthropic found itself at the center of a massive scandal. Immediately after implementing a new feature of "prominent watermarks" in all texts created or edited with the help of the AI assistant Claude, the company faced a sharp negative reaction from the community. The new measure was introduced as part of compliance with the European Union's Artificial Intelligence Act, aimed at transparently distinguishing between human-created content and machine-generated content.

However, the implementation of the mechanism proved to be highly controversial. Watermarks are embedded in the text at the level of metadata and visual markers which, according to users, cannot be ignored when publishing materials. This decision affected a wide range of professionals: from journalists and copywriters to programmers and academic researchers, whose work requires strict adherence to copyright and corporate standards.

Criticism from the professional community

The main criticism focused on the fact that the labeling system does not distinguish the degree of AI involvement in creating the text. Professional developers and editors report that labels persist even with minimal editing, local translation, or using the model solely for grammar checking. In conditions of fierce competition and strict contracts, where content originality is a key requirement, the presence of an AI "stigma" makes using Claude impossible for many corporate clients.

The situation is felt particularly acutely in the academic environment and among freelancers. Students and researchers fear that using Claude to edit essays or scientific papers will automatically lead to accusations of plagiarism or academic dishonesty, as the system does not provide a way to "clean" the text of labels after manual refinement.

Mass user exodus and economic consequences

The market reaction was instantaneous. According to data collected by analysts, there is a significant increase in the number of cancellations of paid subscriptions to the Claude service. Users are mass-migrating to alternative platforms that have not yet implemented such aggressive labeling mechanisms, or are returning to using local models that do not require mandatory identification. For Anthropic, which positioned itself as a leader in "safe" AI, this has become a serious blow to its reputation and financial performance.

Market experts note that mandatory labeling could become a precedent for the entire industry, forcing other AI developers to reconsider their approaches to complying with regulatory norms. However, if the current trend continues, this could lead to market fragmentation and a decrease in trust in generative intelligence tools within the professional environment.

Anthropic's position: defense and explanations

Representatives of Anthropic are trying to smooth over the situation, stating that the watermark is intended only to indicate the fact of AI involvement in text processing, not that the model is the sole author. In their official comments, they emphasize that this is a necessary step to comply with EU regulator requirements, which tightened control over AI content transparency in 2026.

At the same time, the company claims that it does not currently record a sharp overall increase in subscription cancellations, calling the current situation a temporary adaptation of users to new conditions. However, independent sources and reviews on social media indicate the opposite, pointing out that many users have already made the final decision to switch platforms.

Contradictory data

At the moment, there are significant discrepancies in estimates of the scale of the crisis. On the one hand, Anthropic representatives in their official statements (Source ID 2) claim that user churn is not critical and that the company does not see a sharp drop in revenue. They insist that most users understand the necessity of labeling to comply with the law.

On the other hand, independent analytical reports and data from subscription monitoring platforms (Source ID 4) indicate the opposite: there is a surge in cancellations specifically against the backdrop of the introduction of watermarks. Users on social media and specialized forums (Source ID 1) are mass-reporting switching to competitors, calling Anthropic's actions a "tragedy for students and workers." This creates a situation where the company's official position contradicts the real feedback from the community, requiring further monitoring of the situation.