What happened
Google’s Gemini Apps Help page describes a verification tool for signed-in users. A user uploads a single image, video, or audio file and asks a question such as whether it was created or edited by Google AI. The documentation names two mechanisms. SynthID uses invisible watermarks to identify media generated or edited by Google AI models. Content Credentials act as a digital record of content origin and history, including whether AI tools were used; Google describes them as an industry standard developed by the Coalition for Content Provenance and Authenticity (C2PA).
The two mechanisms answer different questions. A detected SynthID watermark means all or part of an image or video was created or edited by Google AI. If no watermark is detected, Google says the media was not created or edited by Google AI, but it could still have come from another AI system. In some cases the result may be inconclusive, for example when there is not enough detail to watermark or an edit was too small. For video, Gemini may identify portions where SynthID was detected; that does not mean every frame in those portions carries a detected watermark. Where credentials are present, Gemini can summarize media composition, edit history, and whether AI was used.
The service is available through gemini.google.com and the Gemini app. Content Credentials verification is available on the web and in the Android app; Google says the iOS app version is coming soon. Users need a personal or qualified Workspace account, must be signed in, and should use the latest Gemini app for in-app verification. A single file can be no larger than 100 MB; videos must be under 90 seconds and audio files under one hour. The page also states a rolling 24-hour usage quota of approximately 10 image checks, 10 video checks (up to five minutes total), and 10 audio checks (up to three hours total).
Google says watermarks usually persist through resizing, recoloring, compression, and other changes, while noting that many alterations can make detection fail. It recommends other checks too, including visual inspection, reverse image search, and metadata review of an original file. The documentation gives no price for the feature.
What the archive adds
This is a new event, distinct from Gemini app lets users share personal avatars for use in friends’ AI images and videos. That earlier note concerns making and sharing media; this update concerns checking provenance. The archive supplies no directly related prior coverage of this verification workflow.
Trend signal
This does not materially fit any supplied SITE_PATTERNS thesis. The feature is a media verification workflow, but the supplied patterns do not address provenance checks or content-authenticity tooling closely enough to support a link.
Why now
The page presents a current, operational path for users to check media inside Gemini, and specifies that Content Credentials verification on iOS is not yet available. It does not explain the timing of the rollout or give a launch date, so there is no stronger timing claim to draw from the source.
Why it matters
For a newsroom, moderation team, or product team reviewing submitted media, the workflow can surface evidence that a file came from Google’s generation tools or expose provenance information attached through Content Credentials. Those are useful signals that can be incorporated into a review process, especially when checking original files rather than screenshots or re-encoded copies.
The constraints matter as much as the feature. A negative SynthID result only rules out a detectable Google watermark under the documented conditions; it does not establish that media is human-made or unedited by other AI systems. Quotas and short maximum video duration also make this an individual checking aid rather than an obvious bulk-ingestion service. Teams should preserve human review and other corroborating methods rather than treating the Gemini response as an authenticity verdict.
Limits and uncertainty
The help page describes capabilities and limits but does not provide measured accuracy, false-positive or false-negative rates, or a validation study. It does not establish how reliably Content Credentials survive transformations, nor whether every file type or provenance format is supported. SynthID coverage is explicitly limited to Google AI tools even though other companies have begun adopting SynthID. The page says watermarks usually persist after common edits but allows that many alterations can defeat detection. It does not specify a price or a programmatic API for this workflow.
Reader takeaway
Gemini now gives signed-in users a concrete way to inspect media provenance, combining Google-specific watermark detection with broader Content Credentials when those records exist. Use it as one evidence source, not a universal AI detector: absence of SynthID says nothing conclusive about content made with other systems, and the checks are quota-limited.
Open question
How broadly will Content Credentials verification become available in the Gemini app, including iOS?