Google’s new AI model Gemini 2.0 Flash is too Good in removing watermarks from images ??

Users on social media have discovered a controversial use for Google’s latest AI model, Gemini 2.0 Flash: removing watermarks from images, including those from major stock media platforms like Getty Images.

Last week, Google expanded access to Gemini 2.0 Flash’s image generation and editing capabilities. While the tool is powerful, it appears to have minimal safeguards. Reports indicate that it can generate images of celebrities and copyrighted characters—along with the ability to erase watermarks from existing photos.

Several users on X and Reddit noted that not only does Gemini 2.0 Flash remove watermarks, but it also fills in the gaps left behind with surprising accuracy. While other AI tools offer similar features, Gemini 2.0 Flash stands out for its effectiveness—and the fact that it’s free to use.

Currently, the image generation feature is labeled as “experimental” and “not for production use.” It is available through developer tools like AI Studio. However, the model isn’t flawless; it struggles with semi-transparent watermarks and those covering large areas of an image.

Despite these limitations, the model’s ability to remove watermarks is likely to concern copyright holders. Competing AI models, such as Anthropic’s Claude 3.7 Sonnet and OpenAI’s GPT-4o, explicitly refuse to remove watermarks, with Claude describing the act as “unethical and potentially illegal.”

Under U.S. copyright law, removing a watermark without permission is generally illegal, barring rare exceptions.

In response to concerns, a Google spokesperson issued the following statement:

“Using Google’s generative AI tools to engage in copyright infringement is a violation of our terms of service. As with all experimental releases, we’re monitoring closely and listening for developer feedback.”

While Google continues to refine its AI models, the debate over ethical AI use—and its legal implications—remains ongoing.