What does Azure AI Content Safety's image scanning look for?
Why C is the answer
Sexual content, violence, hate and self harm
Microsoft lists these categories, with multiple severity levels.
In detail
- It scans for sexual content, violence, hate and self harm.
- Set thresholds to suit your audience.
Use case
An illustrative scenario, not a real incident.
A community site scans uploaded images. Azure AI Content Safety checks for sexual content, violence, hate and self harm.
See it in different ways
flowchart LR A["User prompt"] --> C["Prompt Shields"] B["Documents, emails, web pages"] --> C C -->|"Attack found"| D["Block"] C -->|"Clean"| E["Model"]
sequenceDiagram participant U as User participant D as Uploaded content participant S as Prompt Shields participant M as Model U->>D: Uploads a file or image D->>S: Hidden instructions S-->>M: Blocks or marks the attack
mindmap
root(("Content Safety"))
n1["Harm categories"]
n2["Sexual content"]
n3["Violence"]
n4["Hate"]
n5["Self harm"]Key point: Image scanning looks for harmful content.
Why the other options are wrong
- A. The image's file size and format
- Size and format are not its purpose.
- B. Whether the image has been saved in a search index
- Index membership is not its purpose.
- D. Spelling mistakes in the caption
- Spelling is not its purpose.