2026.06.08 2026.06.08

AWS AI Practitioner AIF-C01: Complete Guide — 5 Domains and a 4-Week Roadmap

swiftwand

The AWS Certified AI Practitioner (AIF-C01) is the newest Foundational-level exam in the AWS certification portfolio, and it certifies a shared vocabulary for the generative-AI era. Can you map AWS AI services such as Amazon Bedrock and Amazon SageMaker AI onto real business problems? Can you explain the decision criteria behind prompt engineering and Retrieval-Augmented Generation (RAG)? This guide brings the exam specification, the weighting of the five domains, the relationship with CLF-C02 and SAA-C03, and a four-week roadmap to a passing level together in a single article.

This AIF-C01 guide is the first step toward AI-focused certification that we previewed at the end of our AWS SAA-C03 Strategy guide.

Why AWS AI Practitioner matters now — the shared language of the GenAI era
The exact numbers of the AIF-C01 exam
Relationship with CLF-C02 and SAA-C03 — where they overlap and what is new
The five-domain map — weightings and themes
The feel of Domains 1 and 2 — the front 44% that maps the AI landscape
The largest weight, 28% — Domain 3 decides pass or fail
Preparing for the new question formats — ordering and matching
Choosing study resources — a three-tier approach starting from free official material
A four-week study roadmap
AIF-C01 for makers and AI engineers
Summary — three things to do today
References

忍者AdMax

Why AWS AI Practitioner matters now — the shared language of the GenAI era

In 2026, the bottleneck in enterprise AI adoption is rarely the technology itself. It is the gap in shared understanding between the people who decide, the people who build, and the people who use AI. AIF-C01 sits exactly on that gap. It does not ask you to train models or tune hyperparameters; it asks whether you can judge when AI is the right tool, which AWS service fits a use case, and how to deploy it responsibly. That judgment is precisely what teams need most as generative AI moves from pilot to production.

The exact numbers of the AIF-C01 exam

Item	Specification
Exam code	AIF-C01
Level	Foundational
Questions	65 (50 scored + 15 unscored)
Duration	90 minutes
Price	$100 USD
Passing score	700 (scaled 100–1000)
Languages	English, Japanese, Korean, Portuguese, Simplified Chinese
Delivery	Pearson VUE test center / OnVUE online proctoring
Recommended experience	Up to ~6 months exposure to AI/ML on AWS
Validity	3 years

Pay attention to the pace. Sixty-five questions in 90 minutes is roughly 1.4 minutes per question. That feels the same as CLF-C02, but the newer question formats (ordering and matching) carry more reading load, so the practical margin is thinner than CLF-C02. Because only 50 questions are scored, a passing performance works out to roughly 70% correct on the scored set.

Relationship with CLF-C02 and SAA-C03 — where they overlap and what is new

Item	CLF-C02	AIF-C01	SAA-C03
Level	Foundational	Foundational	Associate
Duration	90 min	90 min	130 min
Price	$100	$100	$150
Passing score	700	700	720
Core focus	Cloud concepts & core services	AI/ML, generative AI, foundation models	Architecture design
Study time	30–50 h	25–40 h (if CLF done)	80–150 h

If you already hold CLF-C02, the IAM, encryption, and CloudTrail fundamentals tested in AIF-C01 Domain 5 (Security, Compliance, and Governance for AI Solutions) are familiar ground. SAA-C03 tests design judgment around availability, performance, and cost; AIF-C01 tests fit judgment around AI use cases. Distinguishing those two roles is a theme that recurs across every domain.

The five-domain map — weightings and themes

Domain	Title	Weight
1	Fundamentals of AI and ML	20%
2	Fundamentals of Generative AI	24%
3	Applications of Foundation Models	28%
4	Guidelines for Responsible AI	14%
5	Security, Compliance, and Governance for AI Solutions	14%

More than half of the scored content (Domains 2 and 3, 52% combined) is dedicated to generative AI and foundation models. This is the single most important fact for planning your study time: the exam is weighted toward the topics that did not exist on earlier AWS foundational exams.

The feel of Domains 1 and 2 — the front 44% that maps the AI landscape

Domain 1 (20%) establishes vocabulary: the nested relationship of AI, ML, deep learning, and generative AI; supervised, unsupervised, and reinforcement learning; the machine-learning lifecycle; and evaluation metrics such as precision, recall, F1, and AUC. Domain 2 (24%) adds generative-AI fundamentals: tokens, embeddings, chunking, foundation models, and the strengths and limits of generative AI, including hallucination and non-determinism. Together they form the front 44% of the exam — mostly recall, with a layer of judgment.

The largest weight, 28% — Domain 3 decides pass or fail

Domain 3, Applications of Foundation Models, carries the heaviest weight at 28%. It covers model selection, inference parameters such as temperature, RAG with Amazon Bedrock Knowledge Bases, vector databases, the cost gradient of the four customization approaches (prompt engineering, RAG, fine-tuning, and continued pre-training), agents, and prompt-engineering technique. If you are short on time, this is the domain to over-invest in.

Preparing for the new question formats — ordering and matching

Beyond classic multiple choice and multiple response, AIF-C01 uses two newer formats. Ordering asks you to place 3–5 responses in the correct sequence — for example, the stages of the ML lifecycle or the document-to-vector RAG flow. Matching asks you to pair items, such as a service name with its use case, or an evaluation metric with its application. AWS has also been rolling out case-study-style items across its exams, where one scenario carries several questions. Practice the lifecycle and pipeline sequences end to end so ordering questions become mechanical.

Choosing study resources — a three-tier approach starting from free official material

Tier 1 (free, official): the AWS Skill Builder “AWS Certified AI Practitioner” learning plan and the official exam guide PDF. Start here and read the exam guide twice.
Tier 2 (practice): the official AWS practice question set, plus one reputable third-party practice exam to expose yourself to the new formats under time pressure.
Tier 3 (hands-on): a few hours in the Amazon Bedrock console — try a playground prompt, a Knowledge Base, and a model comparison. Touching the services cements Domain 3.

A four-week study roadmap

Week	Focus	Outcome
1	Domain 1 + exam-guide read-through	Vocabulary and lifecycle locked in
2	Domain 2 + Bedrock hands-on	Generative-AI fundamentals and the AWS stack
3	Domain 3 (heaviest) + Domains 4 & 5	Customization, RAG, agents, governance
4	Practice exams + weak-point review	Timed performance above 80%

If you already hold CLF-C02, this compresses comfortably into 25–40 hours. The key is to front-load Domain 3 in week 3 and leave week 4 entirely for timed practice.

AIF-C01 for makers and AI engineers

Even if you build with AI every day, AIF-C01 is worth the four weeks. It forces you to put names to the decisions you already make by instinct — why RAG over fine-tuning here, why a smaller model there, where bias enters a dataset. For a solo maker, that shared vocabulary is what lets you explain your choices to clients and collaborators with confidence.

Summary — three things to do today

Download the official AIF-C01 exam guide and read the five-domain content outline once.
Open the Amazon Bedrock console and run a single playground prompt to make Domain 3 concrete.
Block out four weeks on your calendar, with Domain 3 in week 3 and timed practice in week 4.

AIF-C01 is a 65-question, 90-minute, $100 Foundational certification that proves judgment for the generative-AI era. With most of the weight on generative AI and foundation models, a CLF-C02 holder can reach a passing level in 25–40 hours across four weeks. The next articles in this series break down each domain in detail.