What Is on the AWS AI Practitioner (AIF-C01) Exam? Domains, Format and the October 2026 Update

Answer in brief

AWS Certified AI Practitioner (AIF-C01) is AWS's foundational exam on AI, machine learning and generative AI. It has 65 questions (50 scored, 15 unscored) in 90 minutes. It costs $100 and passes at a scaled 700 out of 1,000.

• What it tests: whether you understand AI, ML and generative AI (GenAI) concepts, can pick the right AWS AI service for a business problem, and can use AI responsibly.
• Domains: five. The largest is Applications of Foundation Models at 28%. Next are Fundamentals of GenAI (24%) and Fundamentals of AI and ML (20%).
• October 2026 update: AWS published version 1.2 of the exam guide on 29 October 2026. AWS says guide updates reach the exam about one month after they are published.
• Agents are in: the current guide covers agentic AI, the Model Context Protocol (MCP), Amazon Bedrock AgentCore and Strands Agents.
• Who it is for: people with up to 6 months of exposure to AI/ML on AWS who use AI solutions but do not necessarily build them.
• How to prepare: read the official exam guide, take a diagnostic by domain, fix the heaviest weak domains first, try Amazon Bedrock hands-on, and book when practice results are strong in every domain.

Everything in this guide comes from the official AWS exam guide and the AWS certification page unless I say otherwise.

The exam at a glance

65 questions, 90 minutes, $100, 700 to pass, no coding.

Four details worth knowing before you book:

1. There are four question types, not two. Ordering questions ask you to put 3–5 steps in the right order. Matching questions ask you to pair 3–7 prompts with answers. You must get every part right to get credit.
2. Unscored questions are hidden. Fifteen of the 65 do not count, but you cannot tell which. Treat every question as scored.
3. Blank answers are wrong. There is no penalty for guessing, so never leave a question empty.
4. Check the language list. The exam is offered in English and several other languages. AWS says the Italian and German versions are discontinued after 15 October 2026.

What changed in October 2026

AWS published version 1.2 of the AIF-C01 exam guide on 29 October 2026. Here is what the official revisions page says, and what it does not.

When it applies to your exam. AWS states that exam guide updates are published approximately one month before they are reflected on the exam. That points to version 1.2 reaching the exam around late November 2026. AWS does not give an exact switch-over date on the revisions page, so treat that as an estimate. If you sit the exam from now on, study to version 1.2.

What AWS lists as changed. The revisions page names one specific change: six services were removed from the out-of-scope list. They are Amazon Elastic Transcoder, AWS Elemental MediaStore, AWS IoT 1-Click, AWS IoT Analytics, AWS IoT Events and Amazon Nimble Studio. They were not added to the in-scope list, so you do not need to study them.

What the current guide covers on agents. You may have read that the October update "added" agentic AI, MCP and Amazon Bedrock AgentCore. All three are in the current (version 1.2) guide. But the AWS revisions page does not say which version first added them, so I will not claim they are new in 1.2. What matters for your exam is that they are in scope now:

• Agentic AI concepts (Domain 2): multi-agent system patterns, the Model Context Protocol (MCP) and its role in connecting agents to external systems, multi-agent communication patterns, memory management, tool usage and workflow orchestration.
• Agentic AI as a term and use case (Domain 1): you must tell agentic AI apart from AI, ML, GenAI and deep learning.
• Agent services (Domain 2): Amazon Bedrock AgentCore and Strands Agents, alongside Kiro and Amazon Quick.
• Agent security (Domain 5): Amazon Bedrock AgentCore Identity and Policy in AgentCore.
• Agent evaluation (Domain 3): evaluating applications built with foundation models, including RAG, agents and workflows.

The domain names and weights in the current guide are the same five listed below.

The five domains and their weights

Domain 3 is the largest at 28%; Domains 2 and 3 together are just over half the scored content.

The weights are AWS's. The "rough scored questions" column is my own arithmetic (weight × 50), not an AWS figure. Use it to plan effort, not as a count.

The services named below are the examples AWS gives in each domain. AWS says the guide is not a complete list of exam content, so treat them as the core, not the edge.

Domain 1: Fundamentals of AI and ML (20%)

What it tests: the basic vocabulary of AI and where it fits. Three tasks:

• Explain basic AI concepts and terminologies
• Identify practical use cases for AI
• Describe the AI/ML development lifecycle

Topics and services that matter: AI, ML, deep learning, neural networks, computer vision, NLP, LLMs, GenAI and agentic AI. Types of inferencing (batch, real-time, asynchronous, serverless). Supervised, unsupervised and reinforcement learning. Regression, classification and clustering. MLOps basics and model metrics such as accuracy, precision, recall and F1 score. Services: Amazon SageMaker AI, Amazon Transcribe, Amazon Translate, Amazon Comprehend, Amazon Lex, Amazon Polly, Amazon Bedrock, Amazon Quick and Kiro.

Expect questions on when AI is not the right answer, and when a traditional ML model beats a foundation model (for example, when explainability or regulation matters).

Domain 2: Fundamentals of GenAI (24%)

What it tests: how generative AI works, what it is good and bad at, and which AWS services build it. Three tasks:

• Explain the basic concepts of generative AI
• Understand the capabilities and limitations of GenAI for solving business problems
• Describe AWS infrastructure and technologies for building GenAI applications

Topics and services that matter: tokens, chunking, embeddings, vectors, prompt engineering, transformer-based LLMs, multi-modal and diffusion models. The foundation model lifecycle. Token-based pricing. Context engineering. Agentic AI concepts, including MCP. Hallucinations and nondeterminism. Services: Amazon Bedrock, Amazon SageMaker AI, SageMaker JumpStart, Amazon Quick, Kiro, Strands Agents and Amazon Bedrock AgentCore.

This is where agents live. Know what MCP is for, and what a multi-agent system does, at a concept level.

Domain 3: Applications of Foundation Models (28%)

What it tests: designing, prompting, customising and evaluating applications built on foundation models (FMs). Four tasks:

• Describe design considerations for applications that use FMs
• Choose effective prompt engineering techniques
• Describe the training and fine-tuning process for FMs
• Describe methods to evaluate FM performance

Topics and services that matter: FM selection criteria (cost, modality, latency, model size, prompt caching). Inference parameters such as temperature. Retrieval Augmented Generation (RAG) with Amazon Bedrock Knowledge Bases. Vector stores: Amazon OpenSearch Service, Amazon Aurora, Amazon Neptune and Amazon RDS for PostgreSQL. Customisation trade-offs: pre-training, fine-tuning, in-context learning, RAG and model distillation. Prompt techniques: zero-shot, single-shot, few-shot, chain-of-thought and prompt templates. Prompt risks: exposure, poisoning, hijacking and jailbreaking. Amazon Bedrock Prompt Management. Evaluation: Amazon Bedrock Model Evaluation, ROUGE, BLEU, BERTScore and LLM-as-a-judge.

This is the heaviest domain. If you only have time to go deep on one, make it this one.

Domain 4: Guidelines for Responsible AI (14%)

What it tests: building AI that is fair, safe and explainable. Two tasks:

• Explain the development of AI systems that are responsible
• Recognize the importance of transparent and explainable models

Topics and services that matter: bias, fairness, inclusivity, robustness, safety and veracity. Legal risks of GenAI, such as intellectual property claims and hallucinations. Bias and variance, overfitting and underfitting. Services: Amazon Bedrock Guardrails, Amazon SageMaker Model Cards and Amazon Bedrock Model Evaluations.

Domain 5: Security, Compliance, and Governance for AI Solutions (14%)

What it tests: securing AI systems and meeting governance rules. Two tasks:

• Explain methods to secure AI systems
• Recognize governance and compliance regulations for AI systems

Topics and services that matter: IAM roles, policies and permissions, encryption, Amazon Macie, AWS PrivateLink, the AWS shared responsibility model, Amazon Bedrock AgentCore Identity, Policy in AgentCore and Amazon Bedrock Guardrails. Prompt injection, data leakage prevention and hallucination detection with RAG grounding. Governance services: AWS Config, Amazon Inspector, AWS Artifact, AWS CloudTrail and AWS Trusted Advisor. The Generative AI Security Scoping Matrix.

What is out of scope

AWS lists job tasks the target candidate is not expected to do, including:

• Developing or coding AI/ML models or algorithms
• Implementing data engineering or feature engineering techniques
• Performing hyperparameter tuning or model optimisation
• Building and deploying AI/ML pipelines or infrastructure
• Conducting mathematical or statistical analysis of AI/ML models
• Implementing security or compliance protocols for AI/ML systems
• Developing and implementing governance frameworks and policies for AI/ML solutions

Note the difference: you must describe security and governance (Domain 5), but you are not expected to implement them.

Who should take it

Anyone who works with AI on AWS but does not build models: analysts, managers, sales and pre-sales, product people, and engineers starting out in AI.

AWS describes the target candidate as having up to 6 months of exposure to AI/ML technologies on AWS. They use, but do not necessarily build, AI/ML solutions.

AWS recommends familiarity with:

• Core AWS services, such as Amazon EC2, Amazon S3, AWS Lambda, Amazon Bedrock and Amazon SageMaker AI
• The AWS shared responsibility model
• AWS Identity and Access Management (IAM)
• AWS service pricing models

There are no required prerequisites. The certification page suggests that people new to IT and AWS take foundational AWS training first. If you already hold AWS Certified Cloud Practitioner or an Associate certification, you can go straight to AI study.

If you are weighing it against a non-AWS AI credential, I compare the two in Claude certification vs AWS AI Practitioner.

How to prepare: a step-by-step plan

Start with a diagnostic by domain, fix the heaviest weak areas first, try Bedrock hands-on, then book on evidence.

1. Read the official exam guide end to end. Make sure it is version 1.2. Every domain page lists tasks and objectives. Use it as a checklist and mark each one as strong, shaky or unknown.
2. Take a diagnostic before any course. Find out which of the five domains are weak. Time spent on what you already know is time taken from a gap.
3. Prioritise by weight × weakness. A weak Domain 3 (28%) costs you twice as much as a weak Domain 4 (14%). But scoring is overall, so do not leave any domain badly weak.
4. Learn the agent vocabulary. Agentic AI, MCP, multi-agent patterns, memory and tool use, AgentCore and Strands Agents. Foundational questions test what these are and when to use them, not how to code them.
5. Get hands-on with Amazon Bedrock. Try a few models in the console, change the temperature, add a guardrail and set up a small knowledge base. The service names stick far better once you have clicked through them.
6. Practise the four question types. Ordering and matching questions punish half-knowledge. Review every wrong answer, write down the domain, and turn each mistake into a flashcard you revisit until it sticks.
7. Book when every domain is consistently strong in practice. My rule of thumb is around 80% per domain, held over several sessions. That is a personal heuristic, not an AWS figure and not a guarantee.

Prepare with StudyTech AI

StudyTech AI is a personalised AWS and AI certification prep platform. It finds your knowledge gaps, builds your practice around your skills and gaps, turns your mistakes into flashcards, scores your readiness for every exam domain, and tells you when you are ready to take your exam.

For AIF-C01, that means you start with a diagnostic across all five domains and spend your time on the weakest ones first.

Pricing is $29 a month or $297 a year, and the diagnostic is free.

• Try 15 free AIF-C01 practice questions, mapped to the five domains with full explanations
• See the AWS AI Practitioner certification page for the full exam overview

Start a free AIF-C01 diagnostic on StudyTech AI

Cite this page

StudyTech AI. "What Is on the AWS AI Practitioner (AIF-C01) Exam? Domains, Format and the October 2026 Update." studytech.ai/blog/aws-ai-practitioner-exam-guide. Updated 8 November 2026.

Related reading

• AI certifications on StudyTech AI
• AWS practice exams on StudyTech AI
• Claude certification vs AWS AI Practitioner
• The best AI certifications: a 2027 tier list

Verified 8 November 2026 against the official AWS Certified AI Practitioner (AIF-C01) exam guide (version 1.2), its five domain pages, the in-scope services list, the revisions page and the AWS certification page. AWS can update the exam guide, so confirm the current version before you book.