AIF-C01 study guide: domains, format and a 4-week plan

AIF-C01, the AWS Certified AI Practitioner exam, checks that you understand AI, machine learning and generative AI concepts and can match AWS AI services to business problems. It is a foundational exam for people who use AI solutions on AWS rather than build them, from line-of-business and IT roles to developers new to the field. You sit 65 questions in 90 minutes, at a Pearson VUE test centre or online.

AIF-C01 at a glance

Details as of September 2026 — confirm on the official exam page before booking.

AWS has announced that the Italian and German versions of the exam retire after 15 October 2026, so candidates planning to sit in either language should check the page before scheduling.

Who this exam is for

AWS describes the target candidate as someone with up to six months of exposure to AI and ML on AWS, who uses AI solutions but does not necessarily build them. The recommended background is familiarity with core services such as Amazon EC2, Amazon S3, AWS Lambda and Amazon SageMaker, the shared responsibility model, IAM, the AWS global infrastructure and AWS pricing models.

There is no prerequisite certification. Just as useful is the guide's list of what is out of scope: coding models, feature engineering, hyperparameter tuning, building pipelines and statistical analysis. If you find yourself memorising algorithm maths or SDK calls, you have gone deeper than the exam goes.

If you already build ML systems on AWS, the [MLA-C02 study guide](/blog/mla-c02-study-guide) covers the associate-level exam, and the [AIP-C01 study guide](/blog/aip-c01-study-guide) covers the professional generative AI developer exam.

What the exam covers

The exam guide has five domains. Domains 2 and 3 together make up more than half the scored content, and both are about generative AI.

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

Expect to distinguish AI, ML and deep learning; supervised, unsupervised and reinforcement learning; batch and real-time inference; and labelled and unlabelled data. The judgement part is picking the right technique (regression, classification, clustering) for a described problem, and recognising when ML is the wrong answer, for example when a rule gives a guaranteed outcome more cheaply. Learn the ML lifecycle stages in order and which SageMaker AI capability belongs to each, plus the difference between model metrics such as accuracy, F1 and AUC and business metrics such as cost per user.

Domain 2: Fundamentals of Generative AI (24%)

Here the exam checks that you understand tokens, embeddings, vectors, chunking, transformers, foundation models, multimodal and diffusion models, and the foundation model lifecycle. It also asks you to weigh generative AI honestly: its strengths, and its weaknesses such as hallucination, nondeterminism and limited interpretability. On the AWS side, know what Amazon Bedrock, SageMaker JumpStart, PartyRock and Amazon Q are each for, and understand the cost trade-offs between on-demand token pricing, provisioned throughput and custom models.

Domain 3: Applications of Foundation Models (28%)

The heaviest domain. It covers how to choose a pre-trained model (cost, modality, latency, language support, input and output length), how inference parameters such as temperature change responses, what retrieval augmented generation is and which AWS databases can store embeddings, and the cost order of customisation approaches from prompting through RAG and fine-tuning to pre-training. It also covers prompt engineering techniques (zero-shot, few-shot, chain-of-thought, templates) and their risks, including exposure, poisoning, hijacking and jailbreaking, and how to evaluate a model with human review, benchmarks and metrics such as ROUGE, BLEU and BERTScore.

Domain 4: Guidelines for Responsible AI (14%)

Expect questions on the dimensions of responsible AI (bias, fairness, inclusivity, robustness, safety, veracity), the legal risks of generative AI such as intellectual property claims, and the effect of bias and variance on different groups. You need to know which tools support responsible AI or document a model, such as Guardrails for Amazon Bedrock and SageMaker Model Cards, and the difference between transparent, explainable models and opaque ones, including the trade-off between interpretability and performance.

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

This domain maps security and governance needs to AWS services: IAM, encryption, Amazon Macie, AWS PrivateLink and the shared responsibility model for protection; AWS Config, Amazon Inspector, AWS Audit Manager, AWS Artifact, CloudTrail and Trusted Advisor for compliance. It also covers data lineage, data governance (residency, retention, logging) and governance processes such as the Generative AI Security Scoping Matrix. Learn each service by its one-line purpose.

A 4-week study plan

The plan below follows the weights: about half your time goes on Domains 2 and 3.

Week 1: Domain 1 and the AWS basics. Build a glossary of the core terms from the guide and write a one-line definition for each in your own words. Map each ML lifecycle stage to its AWS service. Finish with a mixed Domain 1 paper and read every explanation.

Week 2: Domain 2. Study how foundation models work at a conceptual level and what each AWS generative AI service is for. Make a table of generative AI pricing options and when each makes sense. Take a Domain 2 paper at the end of the week.

Week 3: Domain 3. Give this week the most time. Work through model selection, inference parameters, RAG, vector stores, customisation cost order, prompt techniques and evaluation metrics. Take topic papers for each area, then a mixed Domain 3 paper.

Week 4: Domains 4 and 5, then full mocks. Spend the first half on responsible AI and the security and governance services. Then sit a full-length mock under exam conditions, review it thoroughly, fix the weakest topics, and sit a second mock two or three days later.

Where candidates lose marks

• Overthinking a foundational exam. The guide excludes building and tuning models. When two answers look plausible, the one that uses a managed service with the least effort is usually what a foundational question is after.
• Mixing up the customisation ladder. Prompt engineering, RAG, fine-tuning and pre-training differ in cost, effort and what they change. A question about adding current company documents points to RAG; one about changing the model's style or domain behaviour points to fine-tuning.
• Confusing similar AI services. Amazon Comprehend, Transcribe, Translate, Polly, Lex and Textract each do one thing. Learn them by input and output: text analysis, speech to text, translation, text to speech, conversation, document extraction.
• Mismatching governance services. Config records and evaluates resource configuration, CloudTrail records API activity, Artifact provides AWS compliance reports, Inspector scans for vulnerabilities. Many distractors are real services doing the wrong job.
• Partial answers on multiple response, ordering and matching. The guide states you must get every selection, order or pair right to receive credit. Read how many answers are required and check each pair before moving on.
• Learning product names instead of capabilities. Some services the guide names, such as Amazon SageMaker Clarify, are no longer open to new customers. If you understand the capability a question describes (bias detection, drift monitoring, human review of predictions) you can recognise the right answer whatever the service is called.

Practising for AIF-C01 on CertifyAIx

CertifyAIx has 507 original AIF-C01 questions written against the current exam guide, covering every domain and explained option by option. You can read [how our questions are written](/blog/how-certifyaix-practice-questions-are-written) for the details of our method.

• 3 full-length mock exams, 65 questions in 90 minutes, the same as the real exam
• 23 mixed domain papers and 28 topic papers of up to 25 questions each (60 minutes per 25)
• Explanations for every option, with links to the relevant AWS documentation
• Progress tracked by domain and topic

Start with the [AIF-C01 practice questions](/certifications/aws-ai-practitioner-aif-c01).

Frequently asked questions

How difficult is AIF-C01?

It is AWS's foundational AI exam and assumes up to six months of exposure to AI on AWS, with no coding. The challenge is breadth: many terms and services, each tested at the level of what it does and when to use it.

How long should I prepare for AIF-C01?

With some cloud background, three to six weeks of regular study is a realistic range. If AWS or AI terminology is new to you, allow the longer end, because Domains 2 and 3 introduce a lot of vocabulary at once.

Do I need another AWS certification first?

No. There is no prerequisite. Familiarity with core AWS services, IAM and the shared responsibility model helps, and the guide lists these as recommended knowledge.

Are CertifyAIx questions real AIF-C01 exam questions?

No. Every CertifyAIx question is original and written against the public exam guide. Using real exam content breaks the AWS candidate agreement and can cost you the certification.

Is there a penalty for guessing?

No. AWS scores unanswered questions as incorrect and applies no penalty for guessing, so answer every question. Fifteen of the 65 questions are unscored and are not identified.

CertifyAIx is independent and not affiliated with Amazon Web Services. AWS Certified AI Practitioner is a trademark of its owner. All CertifyAIx practice questions are original.