DMR News

Advancing Digital Conversations

Google’s Gemini AI Model Sparks Debate Over Watermark Removal Feature

ByYasmeeta Oon

Mar 19, 2025

Google’s Gemini AI Model Sparks Debate Over Watermark Removal Feature

Users on social media recently discovered a controversial feature in Google’s Gemini AI model: the ability to remove watermarks from images, including those from major stock image providers like Getty Images. The discovery followed the release of Gemini 2.0 Flash’s image generation feature, which allows the model to generate and edit images. This feature enables the AI to erase watermarks and even fill in the gaps left behind, showcasing an impressive level of skill.

The Controversy and Legal Concerns

While other AI tools also offer similar functionality, Gemini 2.0 Flash is being called out for being particularly effective at watermark removal. It can handle both semi-transparent and opaque watermarks, which makes it stand out from other models. Despite its impressive performance, this feature is still in its experimental phase and is only available via Google’s developer-facing tools, such as AI Studio.

The lack of usage restrictions on this tool has raised concerns from copyright holders. Unlike models such as Anthropic’s Claude 3.7 Sonnet and OpenAI’s GPT-4o, which refuse to assist with watermark removal, Gemini 2.0 Flash does not have such limitations. According to U.S. copyright law, removing watermarks without permission is illegal in most cases, which could make this tool a potential facilitator of copyright infringement on a large scale.

In response to the controversy, Google issued a statement reminding users that using its generative AI tools for copyright infringement violates its terms of service. They emphasized that, as with all experimental releases, they are actively monitoring feedback from developers and users to ensure compliance with legal guidelines.

What The Author Thinks

While the advancements in AI are exciting, tools like Google’s Gemini 2.0 Flash bring with them ethical and legal challenges. The ability to remove watermarks with ease could lead to widespread misuse, enabling copyright theft that undermines the efforts of content creators. Google’s lack of strict guardrails and clear restrictions leaves room for significant misuse. Without proper enforcement, AI could become a tool for exploiting intellectual property on a massive scale.


Featured image credit: Wikimedia Commons

Follow us for more breaking news on DMR

Yasmeeta Oon

Just a girl trying to break into the world of journalism, constantly on the hunt for the next big story to share.

Leave a Reply

Your email address will not be published. Required fields are marked *