Home › Resources › Glossary › AI tagging
AI tagging uses artificial intelligence to identify content in a digital asset and add descriptive keywords. In a digital asset management (DAM) platform, it can help marketing teams find images without entering every tag by hand. Tags describe what may be in a file; they do not confirm its usage rights, consent or approval.
HOW IT WORKS
An image-recognition service analyses visual content and returns keywords, such as objects or scenes. Depending on the service, it may also recognise visible text. These keywords become metadata: information stored alongside a file that people can search.
AI tagging is different from copying existing metadata out of a file, or applying a fixed rule based on its folder. It interprets the content, so its results need checking.
Broad visual tags can help someone find a photograph of a bicycle. They may not identify your product model, campaign or approved market. Add that business context using agreed fields and terms. A controlled vocabulary is simply an agreed list of names that everyone uses consistently.
Tagging adds words to an asset record. AI search can match a query to visual content or meaning without relying only on those words. Neither tells you whether a found image is approved for your intended use.
IN PRACTICE
Asset Bank is a DAM platform for marketing, brand and creative teams. Its documented AI auto-tagging options cover objects, text in images, famous landmarks and whether an image contains a face. Detecting a face is not the same as identifying the person or confirming their consent.
Administrators can adjust auto-tagging accuracy settings. Asset Bank also documents separate AI search and prompt-led metadata features, including generated descriptions and alternative text. These are different functions; do not assume an auto-tag is suitable alternative text for a published image.
Check the features enabled for your account and applicable credit limits before planning a large tagging exercise.
AI can mistake an object, miss a detail or use a label that is too broad for your team. A confidence setting can reduce some unwanted suggestions, but it cannot supply missing business context or prove that every retained tag is correct.
Before enabling an external AI service for sensitive imagery, check your organisation’s approved data-handling arrangements and the supplier’s terms. Do not infer sensitive personal characteristics from appearance.
ILLUSTRATIVE EXAMPLE
A communications team uploads a photograph of visitors standing beside a laboratory demonstration. An AI service adds tags such as “people”, “indoors” and “science”. These help describe the scene, but do not identify the event or establish who can appear in a published campaign.
The editor checks the suggestions against the photograph, removes any incorrect labels and adds the verified event name, department and year. They then check the separate consent and usage records before marking the image as available for the planned campaign.
Result: colleagues can search using both visible content and the university’s own terminology. The publishing decision still rests on the recorded permissions, not the AI tags. This is an illustrative workflow, not a customer case study.
AI tagging uses artificial intelligence to identify content in a digital asset and add descriptive keywords. In a digital asset management (DAM) platform, it can help marketing teams find images without entering every tag by hand. Tags describe what may be in a file; they do not confirm its usage rights, consent or approval.
Tagging adds words to an asset record. AI search can match a query to visual content or meaning without relying only on those words. Neither tells you whether a found image is approved for your intended use.
AI can mistake an object, miss a detail or use a label that is too broad for your team. A confidence setting can reduce some unwanted suggestions, but it cannot supply missing business context or prove that every retained tag is correct.
Before enabling an external AI service for sensitive imagery, check your organisation’s approved data-handling arrangements and the supplier’s terms. Do not infer sensitive personal characteristics from appearance.
Explore how Asset Bank helps marketing teams manage, control and share digital assets.
KEEP EXPLORING
