Google has added a visible watermark to videos created in Veo 3

Google has started labeling videos generated using its Veo 3 artificial intelligence model with a special watermark. The announcement was made by Josh Woodward, vice president of Google Labs and Gemini, in a post on social network X.

Watermark in Veo 3 – now visually

The watermark is being added to the visuals of videos generated with Veo 3, except for those created in the Google Flow tool by users with an AI Ultra subscription. The new mark is visible, but it’s hard to notice in normal viewing – it’s embedded into the image with a high degree of care.

Google has previously embedded an invisible digital identifier, SynthID, into its content to identify AI-generated content. Now a visual layer of protection has been added as well. According to Woodward, it’s “the first step toward expanding access to the SynthID detector,” which is still available to a limited number of testers.

Woodward said it’s “the first step toward expanding access to the SynthID detector.”

Veo 3 continues to gain popularity

The Veo 3 video generation model was unveiled at the Google I/O 2025 conference. The system creates highly realistic video with synthesized audio and dialog. Many examples have already appeared online, from fantasy animal sequences to commercials, street interviews, mock news stories, and anboxing videos.

One last thing: We've also added visible watermarks to all videos, except for videos from Ultra members in Flow. This is a first step as we work to make our SynthID Detector available to more people in parallel.

The addition of a visible watermark, according to the developers, is necessary to increase transparency in the distribution of AI content, especially as it grows rapidly.

The story Google adds visible watermark to videos created in Veo 3 was first published on ITZine.ru.