Machine learning fundamentals — AIF-C01
Learn the AI vs. ML vs. deep learning hierarchy for AWS AIF-C01 — definitions, the subset relationship, and the exam trap candidates most often fall into.
WHAT IT IS
Machine learning is a branch of artificial intelligence (AI) and computer science that leverages data and algorithms to enable AI systems to learn and improve. Rather than following predetermined rules, an ML system performs data analysis tasks — such as classifying documents, labeling images, or generating predictions — by finding patterns in data it has been given. Deep learning is a subset of machine learning that uses artificial neural networks to process data in a way inspired by the human brain.
These three terms describe a nested hierarchy: AI is the broadest concept, ML is a subset of AI, and deep learning is a subset of ML.
Mental model
Think of three concentric circles. The outermost circle is AI — the broad goal of making machines behave intelligently. Inside it sits ML — one specific approach to AI that teaches systems to learn from data rather than from hand-written rules. Inside ML sits deep learning — a specialized ML technique that uses layered artificial neural networks to handle complex pattern recognition tasks.
Every deep learning system is also an ML system. Every ML system is also an AI system. The reverse is not true in either direction.
When to use it
The exam tests whether candidates can match a problem to the correct level of the hierarchy and distinguish the three terms from one another.
| Concept | Core idea | Distinguishing trait |
|---|---|---|
| Artificial intelligence | The broad concept of machines performing tasks that require human-like intelligence | Encompasses all approaches — rules-based, ML, and otherwise |
| Machine learning | A subset of AI that uses data and algorithms to learn without explicit programming | Requires training data; the system improves through exposure to examples |
| Deep learning | A subset of ML that uses multi-layered artificial neural networks | Particularly effective on complex, unstructured data such as images, text, and audio |
Use this table when a question asks you to identify which term applies to a described system, or to rank the three from most general to most specific.
COMMON MISCONCEPTION
A common misconception is that "AI," "ML," and "deep learning" are interchangeable synonyms, or that ML and AI are the same thing. They are not. As the official AWS documentation states directly: "machine learning is one of many branches of AI — while machine learning is AI, not all AI activities can be called machine learning." Deep learning, in turn, is a specialized form of machine learning, not a synonym for it. Treating the three as equivalent will lead to wrong answers on questions that ask you to identify which approach is being described in a scenario.
A second trap is assuming that because deep learning is the most capable technique for complex tasks, it is always the appropriate choice. The hierarchy is about scope and mechanism, not about one approach being universally superior. The exam may present a straightforward structured-data problem where classical ML techniques are the described solution — and candidates who default to "deep learning" as the answer will miss the distinction.
How it shows up on the exam
Task Statement 1.1 explicitly calls for candidates to "describe the similarities and differences between AI, ML, and deep learning" — so questions on this concept test recall and discrimination: can you correctly place each term within the hierarchy, and can you match a described system to the right level?
Signal phrases that indicate this concept is in play:
- "Which of the following best describes the relationship between AI, ML, and deep learning?"
- "A system learns to classify images without being explicitly programmed with rules — which term most accurately describes this?"
- "Which statement correctly distinguishes machine learning from artificial intelligence?"
When you see these phrases, anchor to the subset relationship: deep learning is a subset of ML, which is a subset of AI. Candidates often confuse deep learning with ML in general, or treat AI as equivalent to ML. The official AWS documentation is explicit that these are distinct, nested categories — not synonyms.
Related concepts
Sources
Every claim on this page traces to the public exam blueprint and official documentation: