AI-900 Practice Question: Describe features of computer vision workloads on Azure

Answer analysis

Option-by-option breakdown

For each option: why learners choose it and why it is or isn't the right answer here.

• ✗

  Recognising when a web form has been submitted by a user in a browser application

  Why it's wrong here

  Monitoring a web form’s submission is a client- or server-side web development task — for example, using JavaScript event listeners to detect a click on a submit button, or handling an HTTP POST request at an endpoint. Azure AI Document Intelligence processes static images and PDF documents; it has no browser context, session state, or ability to observe user interactions. Therefore, recognizing when a user submits a form in a browser is outside its scope and is not a form recognition capability.

• ✓

  Extracting key-value pairs and tables from structured form documents using pre-built or custom models

  Why this is correct

  Azure AI Document Intelligence (formerly Form Recognizer) uses pre-trained or custom-trained models to locate and extract key-value pairs, tables, and selection marks from structured forms such as tax returns, loan applications, and purchase orders. The OCR and layout analysis layer first reads the text, then the model interprets semantic relationships and returns a structured JSON response containing each field’s extracted value, bounding box, and confidence score. This is exactly the task the service was designed for.

• ✗

  Generating HTML forms automatically from a database schema

  Why it's wrong here

  Generating HTML forms from a database schema is a software scaffolding activity — tools such as Django or ASP.NET can infer input controls from model metadata to emit markup for clients. Document Intelligence performs the opposite transformation: it ingests a document image and emits structured data, rather than consuming data and emitting HTML. The service never generates HTML or user-facing UI; it only reads and understands pre-existing printed or handwritten forms.

• ✗

  Validating that completed forms meet schema and data type requirements

  Why it's wrong here

  Validating extracted data against schema constraints (data types, ranges, required fields) is a downstream step performed after Document Intelligence returns its JSON payload, often using validation rules, JSON Schema validators, or Power Automate conditions. The form recognizer’s output may include confidence scores, but it does not enforce or verify that the values meet business rules; it simply extracts what was written. Consequently, treating validation as the service’s primary function is incorrect, because validation logically depends on extraction first.

Go deeper

Related to this question