Google will now allow users to remove visible watermarks from AI content (www.engadget.com)

Google has announced that it will soon allow users to remove visible watermarks from AI-generated content, a decision that has significant implications for the AI/ML community. This feature will soon be available for models like Nano Banana, Omni, and Lyria, as well as Gemini and the video editor Flow. While the move aims to enhance creative flexibility for users, it undermines the primary purpose of visible watermarks, which are designed to help the public easily identify AI-generated media. Importantly, the change does not affect the invisible SynthID watermark and C2PA metadata, which will remain in place for transparency. This development comes amidst a backdrop of increasing regulatory scrutiny regarding AI transparency, particularly within the EU and South Korea, where laws mandate visible watermarks on deepfake content. Google’s approach highlights the ongoing tension between creative freedom and the need for safety mechanisms in the context of AI use. As the landscape evolves, other companies, such as Suno and Anthropic, are also introducing new transparency measures, indicating a broader shift towards balancing creative uses of AI with the imperative to maintain trust in digital content.