You need to pull invoice totals and dates out of scanned PDFs. Which AI workload is this?
Why A is the answer
Information extraction
Getting structured values out of documents is information extraction. In Foundry that is what Content Understanding does.
In detail
- Pulling fields from documents is information extraction.
- Name the input and output to identify a workload.
Use case
An illustrative scenario, not a real incident.
An accounts team receives scanned invoices. Information extraction pulls totals and dates into fields, so staff stop typing them by hand.
See it in different ways
flowchart TD
A{"What is the input?"} -->|"Text"| B["Text analysis"]
A -->|"Audio"| C["Speech recognition"]
A -->|"Image"| D["Computer vision"]
A -->|"Documents"| E["Information extraction"]
A -->|"A prompt, wanting new content"| F["Generative AI"]mindmap
root((AI workloads))
Text
Sentiment
Key phrases
Entities
Speech
Recognition
Synthesis
Vision
Classification
Object detection
OCR
Documents
Information extraction
Generative
Text
Imagesflowchart LR S0["Scanned PDF"] --> S1["Information extraction"] --> S2["Totals and dates"] --> S3["Accounts system"]
Key point: Fields from documents means information extraction.
Why the other options are wrong
- B. Speech synthesis
- Speech synthesis creates audio.
- C. Image generation and editing
- Image generation creates images.
- D. Sentiment analysis and summarization
- Sentiment analysis reads opinion, not fields.