Digital watermarking
Digital watermarking is the embedding of visible or invisible identifying information into digital content to indicate, verify, protect, or trace its origin or ownership.
Introduction to Generative AI, Second Edition
A watermark is a visible or embedded mark placed in an image or other media to indicate ownership, origin, or authenticity.
- How it is used
- Uses watermarking to identify AI-generated media through signals embedded at creation rather than relying solely on later detection.
- How deep it goes
- Explains image and text watermarking intuitively, including vulnerabilities, without providing an implementation procedure.
- Examples
- Illustrates image watermarking with SynthID and text watermarking with SynthID-Text, and discusses alleged Getty watermark reproduction in Stable Diffusion outputs.
- What you can do after
- Evaluate watermarking as one verification measure by recognizing editing, evasion, adoption, and quality limitations.
Read how Introduction to Generative AI, Second Edition explains digital watermarking →
How Large Language Models Work
Watermarking is the process of embedding an identifiable signal or alteration in content to indicate its origin or authenticity.
- How it is used
- Uses identifiable alterations to distinguish generated content and reduce its unnoticed reuse in training data.
- How deep it goes
- Defines watermarking and gives a homoglyph example, but stops short of specifying a robust image-watermarking algorithm.
- Examples
- Uses subtle output changes and homoglyphs as examples, with teachers needing more reliable generated-content identification.
- What you can do after
- Evaluate watermarking as a practical identification aid while recognizing uncertainty about how reliably it works.
Read how How Large Language Models Work explains digital watermarking →
Also covered in
What to know first
- false positiveA false positive is a positive result that incorrectly indicates a target condition is present when it is absent.
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- Stable DiffusionStable Diffusion is a family of latent diffusion models that generates and edits images from text and other conditioning inputs.
Related concepts
- provenancealternative to · Provenance is traceable information about the origin and history of data or artifacts, including their derivation, transformations, and movement.
- response generationuses · Response generation is the computational process by which a language model produces a natural-language answer or other output conditioned on an input, instructions, or context.
- SynthIDincludes · SynthID is Google's technology for embedding and detecting imperceptible digital watermarks in AI-generated content.
- data loss preventionused by · Data loss prevention is the use of policies and controls to detect and prevent unauthorized disclosure, transmission, extraction, or loss of sensitive data.
- noiseaddresses · Noise is variation, information, or content that is random, irrelevant, or unwanted relative to a signal, pattern, measurement, or task.
- homographuses · A homograph is a word that has the same spelling as another word but a different meaning, and sometimes a different pronunciation.
Questions about digital watermarking
What is digital watermarking?
Digital watermarking is the embedding of visible or invisible identifying information into digital content to indicate, verify, protect, or trace its origin or ownership.
Which Manning books explain digital watermarking?
Introduction to Generative AI, Second Edition and How Large Language Models Work.
What should I know before learning digital watermarking?
Manning books that teach digital watermarking usually introduce false positive, language model and Stable Diffusion first.
What is the difference between digital watermarking and provenance?
Digital watermarking is the embedding of visible or invisible identifying information into digital content to indicate, verify, protect, or trace its origin or ownership. Provenance is traceable information about the origin and history of data or artifacts, including their derivation, transformations, and movement.
What kinds of digital watermarking are there?
Kinds of digital watermarking include SynthID.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

