Hello, you are using an old browser that's unsafe and no longer supported. Please consider updating your browser to a newer version, or downloading a modern browser.
Everything you need to know about the AWS AI Practitioner, the foundational AI certification from Amazon Web Services and the entry point to the AWS AI and machine learning track, as of September 2026. This guide covers the AIF-C01 exam, the five domains and their weights, the question types, the exam format and passing score, DoD 8140 status, the roles it fits, and how it compares with the Cloud Practitioner and the Solutions Architect Associate.
The AWS Certified AI Practitioner is Amazon Web Services' foundational AI certification. It is the AWS exam written for people who need to understand artificial intelligence, machine learning and generative AI on AWS without being tied to a technical build role.
It validates an overall knowledge of the field: how AI, ML and deep learning relate, where a prediction adds value and where it does not, what foundation models are and how services such as Amazon Bedrock, SageMaker and Amazon Q put them to work, how to prompt and evaluate a model, how to use AI responsibly, and how to secure and govern an AI solution. Coding a model, feature engineering, hyperparameter tuning, building pipelines and the math behind a model are explicitly out of scope.
The credential carries no prerequisites. AWS describes the target candidate as someone with up to six months of exposure to AI and ML technologies on AWS who uses but does not necessarily build AI solutions, and suggests that anyone new to IT take the Cloud Practitioner first. It is issued and maintained by Amazon Web Services, and the current exam version is AIF-C01.
AI, ML and deep learning concepts, practical use cases and the ML development lifecycle. 20%.
Foundation model concepts, business capabilities and limits, and the AWS generative AI services. 24%.
Model selection, RAG, agents, prompt engineering, fine-tuning and evaluation. The largest domain at 28%.
Bias, fairness, safety, legal risk, and transparent and explainable models. 14%.
Securing AI systems with AWS controls, and the compliance and governance frameworks around them. 14%.
Four things that make the AI Practitioner the credential most people start with when they start with AI on AWS.
AWS runs the largest share of the cloud market, and a large share of the generative AI work in enterprises runs through Amazon Bedrock and SageMaker. The AI Practitioner is the foundational rung of the AWS AI and machine learning path. It puts the model vocabulary and the AWS service map in place before the Machine Learning Engineer Associate starts asking you to build with them.
AWS wrote this exam for the people around an AI project as much as the people inside it. Cloud, development, data, IT, AI and line-of-business roles all sit inside the target audience, and the exam guide says the candidate uses but does not necessarily build AI solutions. It is role-independent by design.
Domains 2 and 3 of the AIF-C01 guide, Fundamentals of Generative AI and Applications of Foundation Models, carry 52 percent of the scored content between them: Bedrock, prompting, RAG, agents, fine-tuning and evaluation. Our rundown of every major AI certification in 2026 shows where that focus sits against the other AI credentials.
The AI Practitioner is not a DoD 8140 credential. It does not appear in the Qualification Matrix V2.1, so on its own it will not satisfy a cyber-coded position requirement. The current matrix is published at the DoD Cyber Exchange.
Its weight comes from the market instead. AWS is the dominant cloud platform across enterprise and government, Amazon Bedrock and SageMaker are the services many of those organizations use to run AI, and the exam is built around the AWS Well-Architected principles of security, responsibility and governance that every AI deployment has to answer for. Federal cloud teams ask for AWS certifications even though they sit outside the 8140 matrix.
Everything you need to know about the certification, the AIF-C01 exam, and how to keep it current, as of September 2026.
The AI Practitioner is a starting point on the AWS AI track. From here, most people move up to the Machine Learning Engineer Associate, round out the foundation with the Cloud Practitioner, or step across to the Solutions Architect Associate. The three paths below are the common next moves.
The step up on the AI track. The AWS Certified Machine Learning Engineer Associate (MLA-C01) moves from understanding models to preparing data, training, deploying and monitoring them on AWS, and earning it also renews your AI Practitioner. It is the credential that turns AI fluency into an engineering role.
The companion foundation, taken before or alongside. The Cloud Practitioner (CLF-C02) covers the AWS services, security model and billing the AI exam assumes you already know, in the same 65 question, 90 minute format with the same 700 passing score and $100 fee. AWS recommends it first for anyone new to IT.
The technical step across. The Solutions Architect Associate (SAA-C03) is the most widely held AWS certification: 65 questions in 130 minutes, a 720 passing score and a $150 fee. For an AI Practitioner holder who wants to design the infrastructure that AI workloads run on, it is the natural next exam.
The AI Practitioner sits at the foundational tier of the AWS AI track. It is the first rung, the Associate exams build on it, and from there the path goes deeper into AWS architecture, operations and development.
The AI and cloud groundwork
The technical step up
Pick your path
Two questions to answer before you commit: can you sit it, and should you pursue the AI Practitioner specifically. Here's a straight answer to both.
No prerequisites, no required training, no data science background needed.
The AI Practitioner has no prerequisites at all. You register through AWS Certification, pick a Pearson VUE test center or an online proctored slot, and sit the exam. AWS wrote it for people with up to six months of exposure to AI and ML on AWS who use AI solutions rather than build them, which makes it a credential a product manager, analyst or IT manager can take on equal footing with an engineer. The one caution AWS gives: if you are new to IT, take the Cloud Practitioner first.
The AI concepts are new to most people, and so is the AWS service list.
Nothing on the exam goes deep, but it asks you to hold two sets of vocabulary at the same time. You need the AI side, from embeddings and temperature to RAG, RLHF and ROUGE, and you need the AWS side, from Bedrock, SageMaker JumpStart and Amazon Q to Macie, PrivateLink and Audit Manager. Candidates who know one side well and skim the other stall on the scenario questions that name two similar services and ask which one fits. Read the exam guide's in-scope service list early and work through it deliberately.
The AI Practitioner fits roles that use AI on AWS rather than build the models. On its own it strengthens cloud, data, product and management positions; paired with the Machine Learning Engineer Associate it opens the ML engineering ladder.
Builds and runs AWS environments and is increasingly asked to stand up Bedrock, SageMaker and the vector stores behind them. The AI Practitioner gives an engineer the model vocabulary and the responsible AI checks that the data science team expects them to know.
Works with the data that feeds models and the metrics that judge them. The exam's coverage of data types, ML lifecycle stages, evaluation metrics and business measures maps directly to the analyst's side of an AI project.
Scopes AI work for clients or internal customers. The credential proves a consultant can tell a good generative AI use case from a bad one, explain RAG versus fine-tuning, and put a realistic cost and risk frame around a proposal.
Decides which AI features ship and how. The exam's emphasis on selecting models, weighing hallucination and nondeterminism risk, and measuring business value is the product manager's daily work with a name on it.
Approves, funds and governs AI adoption without building it. Domains 4 and 5 give a manager the responsible AI, security and compliance language to set policy and hold vendors and teams to it.
Integrates foundation models into applications through Bedrock and the AWS SDKs. The AI Practitioner covers prompt engineering, inference parameters, agents and the security risks of prompt injection before the developer writes the first call.
The AI Practitioner is one of two AWS foundational exams, and the Solutions Architect Associate is the most common technical step after either. Here's how the three line up.
| AWS AI Practitioner | AWS Cloud Practitioner | AWS Solutions Architect Associate | |
|---|---|---|---|
| Issuer | AWS | AWS | AWS |
| Level | Foundational | Foundational | Associate |
| Focus | AI, ML, generative AI and the AWS AI services | AWS concepts, security, services and billing | Designing AWS architectures |
| Prerequisites | None (up to 6 months AI/ML exposure) | None (up to 6 months exposure) | None (rec. 1 year) |
| Exam | 65 items, 90 min (AIF-C01) | 65 items, 90 min (CLF-C02) | 65 items, 130 min (SAA-C03) |
| Domains | 5 | 4 | 4 |
| Passing Score | 700 of 1000 | 700 of 1000 | 720 of 1000 |
| Exam Cost | $100 | $100 | $150 |
| Validity | 3 years | 3 years | 3 years |
| Good Fit For | People who use AI on AWS in cloud, data, product and business roles | Newcomers and non-technical roles | Cloud architects and engineers |
Pricing varies by region and taxes. The AI Practitioner adds ordering, matching and case study question types to the multiple choice and multiple response formats the other two use. None of the AWS certifications appear on the DoD 8140 Matrix; their recognition is market-driven rather than DoD-listed. Cloud Practitioner and Solutions Architect facts reflect the Training Camp guides for those exams as of September 30, 2026.
Our AWS AI Practitioner boot camp covers all five AIF-C01 domains in one day of live online training, with official AWS courseware, hands-on Amazon Bedrock labs, a mock exam, your $100 exam voucher and a free retake included, for $995.
What moved in the AWS catalog this year, how the AI Practitioner sits against the other AI credentials, and how to study from the exam guide itself.
The 2026 reshuffle of the AWS catalog: which exams retired, which arrived, and where the AI Practitioner sits in the new lineup for someone starting on AWS this year.
The AI credential map from the vendor foundational exams to the governance and security certifications, with a straight read on which one fits which job.
The companion foundational exam. Who gains from it, who can skip it, and why AWS suggests it before the AI Practitioner for anyone new to IT.
The multi-cloud question from the AWS side. How the two platforms divide the market, what a second cloud adds to a resume, and when it is worth the extra exam.
The AIF-C01 exam guide lists every task statement AWS can test. This is how to turn that document into a study plan instead of skimming it once.
A sober read of the certification salary charts, why foundational credentials sit where they do, and what the numbers reflect about experience and role.
How each certification body keeps a credential current, from AWS's three year recertification cycle to the annual CPE models used by ISC2 and ISACA.
The AIF-C01 exam is organized into five domains, each carrying its own weight. Applications of Foundation Models is the largest at 28 percent, Fundamentals of Generative AI follows at 24 percent, and together the two generative AI domains cover more than half of the scored content. Click any domain for what it covers.
The vocabulary and the working parts: what separates AI, machine learning and deep learning, the difference between supervised, unsupervised and reinforcement learning, the data types a model consumes, and when a prediction is the wrong tool for the job. It also walks the ML lifecycle from data collection through training, evaluation, deployment and monitoring, names the SageMaker features and managed services at each stage, and covers the accuracy, AUC and F1 metrics alongside the business measures that decide whether a model was worth building.
Tokens, embeddings, vectors, transformers, diffusion models and the rest of the generative AI vocabulary, plus the lifecycle of a foundation model from data selection and pre-training through fine-tuning, evaluation and deployment. The domain then turns practical: the trade-offs of generative AI for a business problem, hallucinations and nondeterminism included, and the AWS services built for it, namely Amazon Bedrock, SageMaker JumpStart, PartyRock and Amazon Q, along with their token-based pricing and cost considerations.
The largest domain. How to choose a pre-trained model on cost, modality, latency and size, what temperature and output length do to a response, what Retrieval Augmented Generation is and which AWS databases store the embeddings behind it, and where Agents for Amazon Bedrock fit in multi-step tasks. It also covers prompt engineering techniques such as zero-shot, few-shot and chain-of-thought, the risks of prompt injection and jailbreaking, the fine-tuning and RLHF process, and the ROUGE, BLEU and BERTScore metrics used to judge model output.
The features of a responsible AI system, bias, fairness, inclusivity, safety and veracity among them, and the AWS tools that help you check for them: Guardrails for Amazon Bedrock, SageMaker Clarify, SageMaker Model Monitor and Amazon Augmented AI. It also covers the legal risks of generative AI, the characteristics of a well-built dataset, the effects of bias and variance, and why transparent, explainable models documented with SageMaker Model Cards and built with human-centered design matter.
Securing an AI system with the standard AWS controls: IAM roles and policies, encryption at rest and in transit, Amazon Macie, AWS PrivateLink and the shared responsibility model, plus data lineage, source citation and secure data engineering practices. The governance half covers ISO and SOC standards, algorithm accountability laws, the AWS Config, Audit Manager, Artifact, CloudTrail and Trusted Advisor services, data residency and retention, and frameworks such as the Generative AI Security Scoping Matrix.
Domains and weights reflect version 1.4 of the AWS Certified AI Practitioner (AIF-C01) exam guide as published by AWS and verified on September 30, 2026. Confirm the current exam guide on aws.amazon.com before scheduling.
The questions candidates ask most often when researching the AWS Certified AI Practitioner.
The AWS Certified AI Practitioner is the foundational-level AI certification from Amazon Web Services. It validates an overall knowledge of AI, machine learning and generative AI concepts and the AWS services that deliver them, independent of any specific job role: what the technologies are, where they fit and where they don't, how foundation models are applied through services like Amazon Bedrock, how to use AI responsibly, and how to secure and govern an AI solution on AWS. It's written for people who use AI solutions at work rather than build them.
The current exam is AIF-C01, and it's the active version as of September 2026. This page follows version 1.4 of the AIF-C01 exam guide published by AWS. Study from that guide rather than third-party summaries, since it lists every task statement AWS can test and the in-scope and out-of-scope services. Confirm the current guide on aws.amazon.com before you schedule, since AWS revises exam guides between versions.
No. There are no prerequisites and no required training. AWS describes the target candidate as someone with up to six months of exposure to AI and ML technologies on AWS who uses but does not necessarily build AI solutions, and it recommends familiarity with core services such as EC2, S3, Lambda and SageMaker, the shared responsibility model, IAM, the global infrastructure and AWS pricing. If you're new to IT altogether, AWS suggests taking the Cloud Practitioner first.
The five domains are Fundamentals of AI and ML (20 percent), Fundamentals of Generative AI (24 percent), Applications of Foundation Models (28 percent), Guidelines for Responsible AI (14 percent), and Security, Compliance, and Governance for AI Solutions (14 percent). The two generative AI domains together make up more than half of the scored content, so foundation models, prompting, RAG and Amazon Bedrock deserve the largest share of your study time.
AIF-C01 has 65 questions delivered over 90 minutes. Fifty of them are scored and 15 are unscored items AWS uses to evaluate future questions; you can't tell which are which, so treat every question as if it counts. Unanswered questions are marked wrong and there's no penalty for guessing, so answer everything.
Five types. Multiple choice, with one correct answer out of four. Multiple response, with two or more correct answers out of five or more, all of which you must select for credit. Ordering, where you place three to five steps in the correct sequence. Matching, where you pair a list of responses with three to seven prompts and must get every pair right. And case study, where one scenario carries two or more questions that are each scored separately. The ordering, matching and case study formats are newer for AWS exams, so practice reading them carefully.
You need a scaled score of 700 on a scale of 100 to 1,000. Scoring is compensatory, which means there's no minimum per domain: a strong showing in one area can make up for a weaker one. Your report shows a pass or fail result along with performance classifications for each section, so you can see where you were strong and where you weren't.
The AIF-C01 exam costs $100 USD, with variations by region and local taxes. It's delivered at Pearson VUE test centers or as an online proctored exam. Training Camp's AWS AI Practitioner boot camp includes the $100 exam voucher and a free retake in the $995 program fee.
No. AWS certifications don't appear on the DoD 8140 Approved Qualifications Matrix V2.1, so the AI Practitioner on its own won't satisfy a cyber-coded position requirement. Federal agencies and contractors still ask for AWS certifications on cloud and AI roles because AWS is the dominant cloud platform in government, but that demand comes from the market, not the matrix. See the full DoD 8140 work role paths.
The certification is valid for three years. AWS gives you two ways to renew: retake the current AI Practitioner exam, or earn the AWS Certified Machine Learning Engineer Associate, which renews the foundational AI credential at the same time. People who stay on the AI track usually renew by moving up rather than retaking.
The AI Practitioner fits roles that use AI on AWS rather than build the models: Cloud Engineer, Data Analyst, Solutions Consultant, Product Manager, IT Manager and Developer. It shows an employer that you can hold a credible conversation about foundation models, RAG, responsible AI and the Bedrock and SageMaker toolset. For roles that train, tune and deploy models, employers look to the Machine Learning Engineer Associate, and the AI Practitioner is the step that gets you there.
If you're new to IT or to AWS, AWS itself suggests the Cloud Practitioner first, because the AI Practitioner assumes you already know what EC2, S3, Lambda, IAM and the shared responsibility model are. If you already work with AWS, or your job is specifically about AI adoption, you can go straight to the AI Practitioner, since neither exam is a prerequisite for the other. Both are foundational, both run 65 questions in 90 minutes with a 700 passing score and a $100 fee, and many people take the two together.
No. Developing or coding models, feature engineering, hyperparameter tuning, building ML pipelines and doing the math behind a model are all listed as out of scope in the exam guide. What the exam does expect is fluency: you need to explain the concepts, pick the right technique or AWS service for a scenario, weigh the risks, and know the security and governance controls. Plan on a few weeks of preparation if the AI vocabulary is new to you, less if you already work alongside a data team.
If you're weighing the certification, planning the AWS AI path, or arranging AI training for a team, tell us where you are and we'll help you map out the right next step.