Artificial intelligence overview — AIF-C01

What it is

Artificial intelligence (AI) is a broad field of computer science focused on enabling machines to perform tasks that would otherwise require human-like problem-solving. According to AWS, AI systems harness data to automate decision-making and carry out capabilities such as image recognition, natural language understanding, and content generation.

AI is the overarching discipline. Machine learning and deep learning are distinct subfields that live inside it — not synonyms for it.

Mental model

Think of three nested containers:

Every machine learning system is an AI system, but not every AI system uses machine learning. Every deep learning system is a machine learning system, but not every machine learning system uses deep learning. The containment is strict in one direction only.

When to use it

The exam tests whether candidates can distinguish AI, ML, and deep learning rather than treating them as interchangeable buzzwords.

Term	Scope	What it does	Example
Artificial Intelligence	Broadest	Enables machines to perform human-like problem-solving tasks	Smart assistants, robotic vacuums
Machine Learning	Subset of AI	Learns from data without explicit instructions; identifies patterns and makes predictions	Document classification, maintenance forecasting
Deep Learning	Subset of ML	Uses neural networks to process complex inputs; a specialized form of ML	Image recognition, speech recognition

Use "AI" when describing the general capability or field. Use "ML" when the mechanism is learning from data patterns. Use "deep learning" when the mechanism specifically involves artificial neural networks handling complex inputs.

Common misconception

A persistent trap is treating AI, ML, and deep learning as three names for the same thing, or assuming that all AI is ML. AWS states explicitly that "not all AI activities are machine learning and deep learning." A robotic vacuum cleaner and a rule-based smart assistant are AI systems that do not necessarily involve machine learning. Likewise, machine learning is described as "one among many other branches of artificial intelligence" — it is a branch, not the whole tree.

The reverse error also appears: assuming deep learning is simply "advanced AI" rather than a specific technique nested inside machine learning, which is itself nested inside AI.

How it shows up on the exam

The cognitive target for this concept is recognition and classification — candidates must correctly place a described system into the right tier of the AI/ML/deep learning hierarchy. A question may describe a system that "processes data without explicit instructions" and ask what category best fits, or it may ask what term correctly describes the relationship between two of the three fields.

Candidates who have internalized "AI and ML are basically the same" will select the wrong tier. Signal phrases to watch for include "broader concept," "subset," "branch of," and "specialized form of." Ground your answer in the hierarchical relationship: AI contains ML; ML contains deep learning.

Related concepts

• Deep Learning
• Machine Learning Basics
• Neural Network