Google makes visible AI watermarks optional

Google will let users turn off the visible watermark on AI-generated images, video, and songs in Gemini and Flow, while retaining invisible SynthID marks and C2PA metadata. TechCrunch’s report says the setting is rolling out to Nano Banana, Omni, and Lyria, while Google’s Credentio announcement says its local validator powers nearly 40 conformant Google products and handles assets at tens-of-billions scale. Creative teams should treat visible labels as presentation policy, not provenance: preserve C2PA and SynthID checks in the same media workflow that already learned Google’s Earth rollback lesson.