Google’s Gemini 2.0 AI tool sparks controversy as people use it for watermark removal

Google’s latest AI model, Gemini 2.0 Flash, is facing backlash after users discovered it can be used to remove watermarks from images, including those published by prominent stock photo agencies like Getty Images. This controversial feature raises questions about copyright infringement and the ethics of generative AI.

The Gemini 2.0 Flash model, which is part of Google’s cutting-edge generative AI tools, allows for the creation and editing of images, including the removal of watermarks from existing photos. While this functionality can enhance creative projects, it has also raised concerns over its potential misuse. Several users on platforms like X (formerly Twitter) and Reddit have pointed out that the AI tool not only removes watermarks but also attempts to seamlessly fill in the areas previously covered by the watermark, making the process remarkably efficient.

However, despite its capabilities, Gemini 2.0 Flash is still an “experimental” feature that is primarily available through Google’s developer-facing tools, such as AI Studio. It is not yet intended for widespread use and comes with the caveat that it is “not for production use.” Although the tool’s performance in watermark removal is not flawless—particularly with semi-transparent watermarks or those covering large areas of an image—it has nonetheless generated concerns due to its ease of use and lack of restrictions.

The issue at the heart of the controversy is the legality and ethics of watermark removal. In the U.S., removing a watermark without the original copyright holder’s permission is illegal under copyright law, except in rare circumstances. This has prompted criticism from some quarters, particularly among content creators and intellectual property advocates who fear the tool could encourage widespread copyright violations.

Other AI models, such as Anthropic’s Claude 3.7 Sonnet and OpenAI’s GPT-4o, have implemented strict safeguards against watermark removal, with Claude explicitly stating that such actions are “unethical and potentially illegal.” These measures reflect broader concerns within the AI industry about the responsible use of generative tools.

In response to the controversy, Google issued a statement emphasizing that using its AI tools for copyright infringement violates its terms of service. The company acknowledged that the Gemini 2.0 Flash model is still in the experimental phase and assured that they are closely monitoring feedback from developers. “As with all experimental releases, we’re monitoring closely and listening for developer feedback,” the statement read.