The short version
AIF-C01 is AWS’s foundational exam on AI, machine learning, and generative AI. It tests whether you can explain the main ideas, pick the right kind of AI approach or AWS service for a business problem, and use these tools responsibly and securely. It does not ask you to build or train models. This guide follows AWS exam guide version 1.2, published October 29, 2026. Compared with version 1.1, the weightings and format are unchanged; AWS’s revision history lists only a change to the out-of-scope services list.
What the exam is
The full name is AWS Certified AI Practitioner. AWS rates it as foundational level. The exam guide says the certification focuses on practical business uses of AI, and it lists four things the exam checks:
- Describing AI, ML, and generative AI concepts, both in general and on AWS.
- Spotting where AI/ML and generative AI fit a business problem.
- Choosing the right type of AI/ML technology for a specific use case.
- Using these technologies responsibly.
Think of it as a “which tool, and why” exam. A typical question gives you a short business situation and asks which service, technique, or setting fits. The wrong answers are usually real things that solve a neighboring problem.
Who it is for
AWS describes the ideal candidate as someone who uses AI/ML technology on AWS but does not necessarily build it, with up to six months of exposure. Roles on the certification page include cloud, development, data, IT, AI/ML, and line-of-business staff. The exam guide also lists background knowledge: the core AWS services, the shared responsibility model, IAM, and AWS pricing models. There is no formal prerequisite. The prerequisites page covers what helps.
Format at a glance
| Item | What AWS says |
|---|---|
| Exam code | AIF-C01 |
| Level | Foundational |
| Time | 90 minutes |
| Questions | 65 in total: 50 scored and 15 unscored |
| Question types | Multiple choice, multiple response, ordering, matching |
| Passing score | 700 on a scale of 100 to 1,000 |
| Delivery | Pearson VUE test center or online proctored |
| Languages | 12 listed, including English, French, Portuguese (Brazil), and Spanish (Latin America and Spain) |
| Validity | 3 years |
Details, including scoring and retakes, are on the exam format page.
The five domains and their weightings
The percentages are shares of scored content, taken from exam guide version 1.2 (unchanged from 1.1).
| Domain | Weighting |
|---|---|
| Fundamentals of AI and ML | 20% |
| Fundamentals of GenAI | 24% |
| Applications of Foundation Models | 28% |
| Guidelines for Responsible AI | 14% |
| Security, Compliance, and Governance for AI Solutions | 14% |
Fundamentals of AI and ML
Core vocabulary (AI, ML, deep learning, training, inference, bias, fit), the types of learning, when AI is a good fit and when it is not, the ML lifecycle, and basic performance metrics such as precision, recall, and F1. It also covers managed services such as Amazon Transcribe, Translate, Comprehend, Lex, and Polly.
Fundamentals of GenAI
Tokens, embeddings, chunking, prompts, foundation models, and the foundation model lifecycle. The strengths and weaknesses of generative AI, such as hallucinations. Token-based pricing and the AWS building blocks, with agentic AI concepts such as tools, memory, and the Model Context Protocol.
Applications of Foundation Models
The biggest domain. It covers choosing a model, inference parameters like temperature, retrieval augmented generation (RAG) with vector stores, customization options and their costs, prompt engineering techniques, fine-tuning data preparation, and evaluation metrics such as ROUGE, BLEU, and BERTScore.
Guidelines for Responsible AI
Bias, fairness, inclusivity, and safety, legal risks of generative AI, dataset quality, overfitting and underfitting, and transparency and explainability, with tools such as Amazon Bedrock Guardrails and SageMaker Model Cards.
Security, Compliance, and Governance for AI Solutions
IAM, encryption, Amazon Macie, PrivateLink, prompt injection, data leakage, grounding, and governance services such as AWS Config, AWS CloudTrail, and AWS Artifact.
Official source
For the vendor’s own description, see the AWS AIF-C01 exam guide. It is the final authority if anything here differs. AWS says exam guide updates appear about a month before they show up on the exam, so check the revision history before you book.
More guides for AIF-C01
- AIF-C01 Study Plan: 4-Week and 8-Week
- AIF-C01 Sample Questions With Answers
- Is the AIF-C01 Exam Hard?
- AIF-C01 Renewal and Retakes Explained
- Is AIF-C01 Worth It? An Honest Look
- AIF-C01 Cheat Sheet: Key Facts to Review
- AIF-C01 vs MLA-C02: Which First?
- AWS AIF-C01 vs AI-901: Which First?
Next step
Read the domains explained page to see where your time should go. When you want to check yourself, our AIF-C01 practice tests are five original practice sets of 65 questions each, with results reported by domain.
Frequently asked questions
Does the AI Practitioner certification expire?
Yes. AWS lists a validity of three years. You can renew by passing the latest version of this exam, or by passing the Machine Learning Engineer – Associate or Generative AI Developer – Professional exam, according to AWS’s recertification page.
Do I need to write code or train models?
No. AWS lists developing models, tuning hyperparameters, and building pipelines as tasks outside the scope of this exam. You need to understand them conceptually.
Is the exam only in English?
No. AWS lists twelve languages. It also states that the exam in Italian and German is retired after October 15, 2026, so confirm the current list when you book.