1.2. The Input-Output Framework

💡 First Principle: The type of input data determines which AI workload category applies. Images go to Computer Vision, text goes to NLP, and audio goes to Speech services. This isn't arbitrary—each data type requires fundamentally different mathematical representations.

What breaks without this framework: Imagine trying to memorize which of 50+ Azure AI services handles each scenario. You'd drown in details. But when you realize that INPUT determines the service category, exam questions become simple: identify what goes IN, and you've eliminated 75% of wrong answers immediately.

Think of it like a postal sorting facility. Packages, letters, and postcards go to different machines—not because someone decided arbitrarily, but because each requires different handling equipment. You don't analyze the contents; you classify by container type. That's exactly what this framework does: classify by input, and the service category reveals itself.