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.

• ✗

  AI that processes data in multiple programming languages simultaneously

  Why it's wrong here

  This confuses programming languages with data modalities. Processing code written in Python, C#, Java, or JavaScript is about source code syntax and runtime support, not about perceiving and relating the content of text, images, and audio. A model may be able to translate between programming languages or generate code, but that is a form of text processing, not multimodal understanding. Multimodal AI is defined by the types of input data being fused, not by the programming languages used to build the model.

• ✓

  AI that processes and relates multiple data types (text, images, audio) together

  Why this is correct

  This correctly defines multimodal AI: a single model or model family that ingests tokens from more than one data modality—such as text, images, audio, and sometimes video—and learns to align them in a shared embedding space. Because the model jointly processes these modalities, it can capture cross-modal relationships like matching a text query to an image, answering questions about a picture, or correlating audio with visual events. This fusion of information enables tasks that would be impossible with separate unimodal models, such as visual question answering, image-text retrieval, and audio-visual analysis.

• ✗

  Deploying AI models across multiple Azure regions for global availability

  Why it's wrong here

  This describes an infrastructure and availability strategy, not a model capability. Deploying models to multiple Azure regions improves latency, fault tolerance, and compliance by distributing identical or independent model instances geographically, but it does not change what each model can do. A single-region multimodal model can still process text, images, and audio together, while a multi-region deployment of unimodal models still has no cross-modal understanding. Therefore, global availability is unrelated to the definition of multimodal AI.

• ✗

  Using multiple AI models in sequence where each model processes a different step

  Why it's wrong here

  This describes a sequential or pipeline architecture in which separate models are chained together, each handling one isolated step or one data type. Such an arrangement does not make the system multi-modal because no single model ever learns to relate or fuse the different data types into a shared representation. Even if text, images, and audio are all processed somewhere in the pipeline, the models remain unimodal and interaction between modalities is limited to passing discrete outputs, not true cross-modal reasoning.

Go deeper

Related to this question