Google’s latest AI model, Gemini 2.0 Flash, is under scrutiny after reports surfaced that it can effectively remove watermarks from copyrighted images. While the company has been working on ways to watermark AI-generated content for authenticity, it seems its technology has now become a tool for stripping those same protections away.
Gemini 2.0 Flash—The Ultimate Watermark Eraser?
Users on X and Reddit have discovered that Gemini 2.0 Flash can remove watermarks from stock images—including those from Getty Images—and seamlessly fill in the gaps where the watermark once was. TechCrunch reports that Google’s model is more adept at this than many existing AI-powered tools, making it a powerful, free-to-use option for image editing.
While Google has labeled Gemini 2.0 Flash’s image generation capabilities as “experimental” and “not for production use,” the implications of its abilities have raised concerns across the industry. Under U.S. copyright law, removing watermarks without the original creator’s consent is illegal, making this a potential legal minefield.
The Double Standard: Google’s AI Ethics in Question
Ironically, Google has been at the forefront of watermarking AI-generated content through its SynthID technology. Last month, the company announced that Google Photos would begin using SynthID to label AI-edited images, reinforcing efforts to trace AI-generated content. However, Gemini’s ability to erase existing watermarks contradicts these efforts, making it easier to misappropriate copyrighted material.
Other AI models—such as OpenAI’s GPT-4o and Anthropic’s Claude—explicitly refuse to remove watermarks, labeling the practice as unethical and illegal. Google’s model’s lack of such restrictions could lead to legal consequences for the tech giant.
The Bigger Picture: A Looming Copyright Showdown
The rise of AI-powered editing tools has already put immense pressure on content creators, photographers, and stock media companies. The Coalition for Content Provenance and Authenticity (C2PA), which includes Google, Microsoft, OpenAI, and Adobe, has been working on technical standards to track AI-generated media. However, adoption challenges remain, and Gemini 2.0 Flash’s watermark removal ability further complicates the fight against digital piracy.
What’s particularly alarming is that Gemini removes watermarks and replaces them with a subtle “edited with AI“ mark via SynthID. While this may seem like a workaround, it does little to protect original copyright holders.
Google’s Silence—And the Road Ahead
Google has yet to respond to concerns regarding Gemini 2.0 Flash’s watermark removal capabilities. The model is only available to developers through AI Studio so that the company may refine its safeguards. However, if this feature becomes widely accessible, it could open the floodgates for misuse, leading to an industry-wide battle over copyright protections in the age of AI.
For now, content creators and copyright holders will closely monitor Google’s response to these concerns. If no action is taken, legal challenges could be imminent, and Gemini 2.0 Flash may become one of the most controversial AI tools in history.
