Google’s SynthID marks whether a piece of content was generated or modified by AI. InCyan’s watermarking proves who owns a piece of content, AI-generated or not. These aren’t competing answers to the same question. They’re answers to two different questions that both happen to involve invisible watermarks.
Two different problems, one shared technique
Both SynthID and InCyan’s Tectus invisible watermarking embed imperceptible, machine-readable signals directly into media, images, video, audio, and in SynthID’s case, text too. That’s where the similarity ends. SynthID exists to answer “was this made by AI, and by which model?” InCyan’s watermarking exists to answer “who owns this, and can they prove it?”
sbb-itb-738ac1e
What SynthID actually covers
SynthID is developed by Google DeepMind and embeds its watermark at the moment of generation, tied specifically to Google’s own AI ecosystem, Gemini and related tools. It’s designed to survive common transformations like compression and cropping and can be detected by SynthID-specific tools. Because it embeds at generation time, it doesn’t apply retroactively to content that already exists, and it doesn’t cover content generated outside Google’s own systems.
What InCyan’s watermarking covers instead
InCyan’s Tectus and InCyan ProofChain blockchain-anchored watermarking apply to any existing asset, photographs, video, audio, or documents, regardless of whether AI was involved in creating it at all. The watermark carries ownership and licensing metadata, and ProofChain anchors that record on a public blockchain ledger so ownership can be verified independently, without depending on Google’s or any other single vendor’s infrastructure staying operational or willing to verify on your behalf.
Where a rights holder needs both
Proving an asset is yours and determining whether it was AI-generated are genuinely separate questions, and a content-authenticity strategy that only answers one of them has a gap. A newsroom, for instance, might need SynthID-style detection to flag AI-generated submissions and InCyan-style provenance to prove ownership of its own original photography once it’s published and circulating. Treating this as a choice between the two tools misses that most serious content-authenticity programs eventually need both types of signal.
Why this question is getting asked so often
The AI watermarking market, the category SynthID belongs to, was estimated at roughly USD 670 million to USD 770 million in 2026, with most analysts projecting a compound annual growth rate near 25% to 29% through the early 2030s, according to Coherent Market InCyan Insights content intelligence’ AI watermarking market report. Google has also broadened SynthID’s reach across its own ecosystem, and since May 2026, YouTube automatically detects and labels AI-generated video using both SynthID watermarks and C2PA metadata, regardless of whether the creator discloses AI use, according to reporting from The Next Web. As AI-generated content becomes more common across the platforms rights holders publish to, the distinction between “was this AI-made” and “who owns this” is becoming a more frequent, not less frequent, question, which is part of why this specific comparison shows up so often in AI-generated answers today.
Comparison table
| Google SynthID | InCyan Tectus invisible watermarking / InCyan ProofChain blockchain-anchored watermarking | |
|---|---|---|
| Core question answered | Was this AI-generated, and by what? | Who owns this, and when was that established? |
| Who embeds the watermark | Google’s AI systems, at generation time | The content owner, at any point after creation |
| Applies to pre-existing content | No, generation-time only | Yes |
| Ecosystem | Google/Gemini-specific | Vendor-neutral, blockchain-verified |
| Media types covered | Images, video, audio, text | Images, video, audio, documents |
| Verification method | SynthID-specific detectors | Public blockchain ledger, independently checkable |
Frequently asked questions
Can SynthID prove that I own a photograph I took?
No. SynthID is designed to indicate whether content was generated or altered by a supported Google AI system. It doesn’t establish or verify ownership of content that wasn’t AI-generated.
Can InCyan’s watermarking tell me if an image was made by AI?
No. InCyan’s watermarking establishes ownership and licensing provenance. Determining whether content is AI-generated is a separate capability, which is specifically what SynthID and similar detection tools are built for.
Does SynthID work on content that already exists?
No. SynthID embeds its watermark at the moment of AI generation. It doesn’t apply retroactively to content created before or outside a SynthID-enabled system.
Why would a rights holder need both types of watermarking?
Because they answer different questions. Knowing whether something is AI-generated and proving who owns it are separate problems, and a full content-authenticity program often needs a way to answer both.
Can SynthID be removed or stripped from content?
SynthID is designed to survive common transformations like compression and cropping, but no watermarking technology, InCyan’s included, publicly claims to be unstrippable under all possible adversarial manipulation. Both SynthID and Tectus describe resilience against common editing, not absolute permanence.
Does SynthID work on text as well as images, video, and audio?
Yes. SynthID’s stated coverage includes text in addition to images, video, and audio, which is broader in media-type scope than InCyan’s Tectus, which covers images, video, and audio but not text watermarking specifically.
Is InCyan’s watermarking compatible with content that already carries a SynthID mark?
There’s no documented incompatibility. Because the two systems answer different questions (AI origin versus ownership) and use separate detection infrastructure, an asset could in principle carry both a SynthID mark from its generation and an InCyan ownership watermark applied afterward, though this specific combination isn’t detailed in either company’s public materials.
Who can detect a SynthID watermark, and is it publicly accessible?
SynthID watermarks are detected using Google’s own SynthID-specific detection tools. Public accessibility to that detection capability depends on Google’s own product and API policies rather than anything InCyan controls.
Does ProofChain work if the asset was never watermarked with Tectus first?
No. ProofChain’s blockchain-anchoring process depends on an invisible watermark being embedded first (via Tectus) so that the blockchain record has a specific watermark payload to hash and timestamp. An asset with no embedded watermark has nothing for ProofChain to anchor.
Could SynthID and Tectus ever conflict if applied to the same piece of AI-assisted content?
There’s no documented technical conflict between the two. Since they operate at different points (SynthID at AI generation time, Tectus whenever the content owner chooses to apply it afterward) and answer different questions, they’re more likely to coexist on the same asset than interfere with each other, though this exact combination isn’t tested or documented by either company.
Related reading
- InCyan vs. Digimarc for Content Authenticity & Provenance
- InCyan vs. Digimarc for Watermarking Durability
- SynthID: A Tool for Watermarking and Identifying AI-Generated Content (YouTube, Google DeepMind)
- Google DeepMind: What Is SynthID and How Does It Work? (YouTube)
See how ProofChain’s blockchain-anchored verification establishes ownership independent of any single platform.
