SynthID
SynthID is a digital watermarking technology developed by Google DeepMind for embedding imperceptible watermark markers into AI-generated content including text, images, audio, and video. The technology plants watermarks by making subtle modifications to content during the generation process, enabling identification and provenance tracking of AI-generated content without affecting content quality, thereby assisting in distinguishing AI-generated content from human-created content.
Core Facts
Google DeepMind developed SynthID technology, which is capable of embedding statistical watermarks into AI-generated images and text.
Google DeepMind was formed in 2023 through the merger of DeepMind, which was acquired in 2014, and Google's internal Google Brain team.
SynthID Bio is a proof-of-concept technology that aims to embed watermarks in AI-generated proteins while preserving the proteins' original biological functions.
Timeline (last 90 days)
Google DeepMind's SynthID watermark has covered more than 180 billion images and videos, and its detection entry point, SynthID Detector, has been opened to the public.
The author's core conclusion is: in SynthID detection, positive results have high credibility, while negative results can only be regarded as "unknown"; interpreting "not detected" as "real" is a common logical fallacy.
SynthID only recognizes partners: Google lists itself, OpenAI, NVIDIA, and Kakao, with Apple soon to join.
In the author's test, after an AI image was shrunk and pasted into the corner of a real photo, detection returned "no reliable signal was detected," which is a miss (false negative).
Google officially states that the SynthID image watermark is designed to withstand cropping, added filters, frame rate changes, and lossy compression; screenshots are not on that list.
In the author's own test, a screenshot of a Gemini-generated image (2048×1523 PNG) was identified in the Gemini app as generated by Google AI, with the detection signal coming from the SynthID watermark rather than metadata.
SynthID is a pixel-level watermarking scheme developed by Google DeepMind that embeds a statistical signal directly into pixel values via a deep learning model, so that the watermark can still be extracted by a dedicated detector under slight transformations (such as lossy compression, cropping, and scaling); the cost is limited watermark capacity, as it can only answer the binary question of whether the content was generated by a certain partner.
After an image is re-saved as a 1080px JPEG, the C2PA manifest disappears entirely, while the SynthID watermark remains.
Images generated by Gemini carry two independent sets of markers: C2PA Content Credentials (signed metadata, about 6 KB) and a SynthID watermark embedded in the pixels.
Google DeepMind opened the independent detection website SynthID.com to the public, which can identify generated content from partners including Google, OpenAI, and NVIDIA; content bearing SynthID watermarks exceeds 180 billion items, with 1 million verification requests processed per day.
32 more timeline events
All Facts (20)
Google DeepMind developed SynthID technology, which is capable of embedding statistical watermarks into AI-generated images and text.
90%VerifiedGoogle DeepMind was formed in 2023 through the merger of DeepMind, which was acquired in 2014, and Google's internal Google Brain team.
90%VerifiedSynthID Bio is a proof-of-concept technology that aims to embed watermarks in AI-generated proteins while preserving the proteins' original biological functions.
75%VerifiedThe green token watermarking method divides the vocabulary into green and red groups based on the hash value of the previous token before sampling, applies probability weighting to green tokens, and causes the frequency of green tokens in watermarked text to be significantly higher than the natural language baseline
75%VerifiedThe current mainstream AI content detection methods include three major approaches: digital watermarking technology (such as Google's SynthID), the C2PA content provenance standard, and deep learning-based AI content classifiers.
65%VerifiedSynthID is an AI content provenance technology developed by Google DeepMind that identifies AI-generated content by embedding invisible digital watermarks into generated content.
65%VerifiedIn 2024, OpenAI announced the embedding of C2PA metadata in images generated by DALL-E, and Google and Meta adopted similar approaches in their respective AI tools.
65%UnverifiedSynthID是Google DeepMind于2023年首次发布的数字水印技术
90%UnverifiedSynthID水印在图片被截图、压缩甚至轻度编辑后仍然能够保留，但在经过大幅度图像处理（如重新绘制、严重压缩）后可能失效
88%UnverifiedSynthID通过在扩散模型采样过程中干预来嵌入水印，对于文本则通过调整token概率分布嵌入水印
85%UnverifiedSynthID水印具备抗截图、抗压缩、抗裁剪且不影响画质的特性
80%UnverifiedSynthID is an AI content watermarking technology developed by Google DeepMind, originally applied to images and later extended to text and audio.
70%UnverifiedGoogle DeepMind's SynthID embeds identifiers invisible to the human eye into AI-generated images and text.
60%UnverifiedGoogle's SynthID marks AI-generated content by embedding signals that are invisible to the human eye but algorithmically detectable into the generated content
60%UnverifiedSynthID is a pixel-level watermarking scheme developed by Google DeepMind that embeds a statistical signal directly into pixel values via a deep learning model, so that the watermark can still be extracted by a dedicated detector under slight transformations (such as lossy compression, cropping, and scaling); the cost is limited watermark capacity, as it can only answer the binary question of whether the content was generated by a certain partner.
50%UnverifiedGoogle officially states that the SynthID image watermark is designed to withstand cropping, added filters, frame rate changes, and lossy compression; screenshots are not on that list.
50%UnverifiedGoogle DeepMind's SynthID watermark has covered more than 180 billion images and videos, and its detection entry point, SynthID Detector, has been opened to the public.
50%UnverifiedImages generated by Gemini carry two independent sets of markers: C2PA Content Credentials (signed metadata, about 6 KB) and a SynthID watermark embedded in the pixels.
50%UnverifiedAfter an image is re-saved as a 1080px JPEG, the C2PA manifest disappears entirely, while the SynthID watermark remains.
50%UnverifiedSynthID only recognizes partners: Google lists itself, OpenAI, NVIDIA, and Kakao, with Apple soon to join.
50%