Common questions about Sasha Signatures: how they protect your images and what they embed
What is the Signature? Is it the same as a digital watermark?
What we've described is very similar to classical digital watermarks, but there's more to Sasha Signatures than that. Outside of ensuring that every Signature is unique, we also tie other unique data to the particular image at hand, making it extremely hard to move a signature from one image to another.
Does the image get distorted by encoding a Signature?
While the image is slightly altered as part of embedding the signature, it is done in a way that only minimally interferes with human perception. Technically speaking, embedding a signature typically results in a PSNR loss 47 which is equivalent to saving a JPEG image at 90% quality.
Do you see the images?
Never. Everything runs on-device. Our models identify what to protect without processing the actual content. All that is embedded into the image is a unique randomized numeric value. This identifier can not be traced to image content, or to who added the Signature, or to where it was done.
What if someone finds the signature?
All data is stored with the platform to ensure the highest level of privacy. Should someone manage to find the signature, it will not make sense and even with access to our technology, they would not be able to access related user data.
Can a signature be forged?
No, we don’t believe so: we’ve gone to great lengths testing against “adversarial attacks” where a signature is applied to another image. In each case so far, we’ve been able to detect it.
What about data privacy?
All data is stored with the platform to ensure the highest level of privacy. The signature is simply a reference.
Is it only for commercial photos?
No. SASHA protects the original intention for all photos, whether private or commercial. How that works technically depends on the individual partner platforms and their setup.
We already use hashing. How is it different?
Unlike hashing systems, which require an exact match and are vulnerable to edits, SASHA’s pixel-level protection embeds invisible watermarks that are resilient to manipulation and maintain user privacy. SASHA is part of the StopNCII community.
What is the difference between SASHA and C2PA
C2PA aims to trace media origins using public metadata, while SASHA focuses on protecting images with invisible watermarks to enable prevention at upload. C2PA traces where media comes from via public metadata. No enforcement capability. SASHA embeds protection into the image and enables active blocking. Complementary, not competing. SASHA is a member of C2PA and CAI.
When you do it on a copy level, we assume it will add a lot of extra storage costs?
Technically, we won’t require any additional image storage, as the image with the SASHA Signature embedded is a pure function of (some-id, image). The stored signature is only around 60 bits.
How it might degrade the UX as assumed it would add time to load / processing?
The signature embedding adds milliseconds to the processing time at load (depending on the image size) with no visible impact on the user experience. In practice, this remains well below typical network variance and image delivery times.
More questions we can answer?
Please don’t hesitate to contact us below!
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.