We earn a commission if you buy through links on this page. How we're funded.
Quick answer
The AWS Certified AI Practitioner (AIF-C01) is AWS's gentlest AI exam — foundational level, no coding, and passable in roughly four to six weeks of part-time study even if you have never opened the AWS console for a living. The preparation that works is unglamorous: study the official exam guide's domains, work through AWS's own free training path, and drill practice questions until your scores sit comfortably above the passing line. For that drilling, we would use Udemy’s AWS Certified AI Practitioner AIF-C01 Practice Exams: four full practice exams and no video, bought once. This guide gives you the week-by-week plan, the resources worth your time, and the mistakes that sink first-timers.
Exam AI-900: Microsoft Azure AI Fundamentals is retired. The replacement is Exam AI-901: Microsoft Azure AI Fundamentals, which earns the same Azure AI Fundamentals certification but expects Python and familiarity with REST APIs and SDKs.
See AWS Certified AI Practitioner AIF-C01 Practice Exams on Udemy →
The table below compares 4 certifications on provider, level, realistic time, coding needed and best for.
| Certification | Provider | Level | Realistic time | Coding needed | Best for | Enrol |
|---|---|---|---|---|---|---|
| AWS Certified AI Practitioner (AIF-C01) | AWS | Foundational | ~4–6 weeks of prep | No | The exam this guide prepares you for | AWS → |
| Azure AI Fundamentals (AI-901) | Microsoft | Foundational | ~2–4 weeks of prep | Basic Python | The Microsoft counterpart (replaced AI-900 on 30 June 2026) | |
| AWS Certified Machine Learning Engineer – Associate (MLA-C01 / MLA-C02) | AWS | Intermediate | ~2–3 months of prep | Yes (Python) | The technical next step after AIF-C01 | AWS → |
| AWS Certified Cloud Practitioner (CLF-C02) | AWS | Foundational | ~2–4 weeks of prep | No | Optional general-AWS grounding, not a prerequisite |
Is the AWS AI Practitioner hard to pass?
No — by AWS exam standards it is the accessible one, designed for business and non-engineering roles as much as for technologists. The difficulty is not depth; it is wording. AWS exams test through scenarios ('a company wants to…, which service should it use?'), so the skill being examined is mapping situations to services and concepts, not reciting definitions. People fail AIF-C01 for two reasons: they skim the generative-AI material because it feels familiar from using chatbots, or they never practise the scenario phrasing before meeting it under time pressure. Both are fixable with the plan below.
Our full assessment of whether this credential deserves a place on your CV is in the AWS AI Practitioner review; this guide assumes you have decided to sit it and want to pass first time.
What's actually on the exam?
Study from the official exam guide, not from a course's table of contents. AWS's AIF-C01 exam guide, read on 16 October 2026, splits the scored content into five domains and lists fourteen task statements under them; the table below is that outline, with the weight AWS gives each domain.
The table below lists the five AIF-C01 domains, the share of scored content AWS gives each, and the task statements AWS lists under it.
| Domain | Share of scored content | What AWS says you must be able to do |
|---|---|---|
| 1. Fundamentals of AI and ML | 20% | Explain basic AI concepts and terminology; identify practical use cases for AI; describe the AI/ML development lifecycle |
| 2. Fundamentals of generative AI | 24% | Explain the basic concepts of generative AI; understand its capabilities and limitations for solving business problems; describe AWS infrastructure and technologies for building generative-AI applications |
| 3. Applications of foundation models | 28% | Describe design considerations for applications that use foundation models; choose effective prompt-engineering techniques; describe the training and fine-tuning process; describe methods to evaluate foundation-model performance |
| 4. Guidelines for responsible AI | 14% | Explain the development of responsible AI systems; recognise the importance of transparent and explainable models |
| 5. Security, compliance and governance for AI solutions | 14% | Explain methods to secure AI systems; recognise governance and compliance regulations for AI systems |
Three things in the same guide change how you should prepare. First, applications of foundation models is the biggest single domain, and with generative-AI fundamentals it makes up more than half the scored content: this is where AWS services such as Amazon Bedrock meet general concepts, the intersection candidates prepare least for. Second, only 50 of the questions are scored; the other 15 are questions AWS is trialling, and nothing on the exam tells you which they are. Third, results are a scaled score from 100 to 1,000 with 700 to pass, and the scoring is compensatory: you pass on the exam as a whole rather than domain by domain, so a weak domain costs you marks without failing you on its own. The questions come as multiple choice, multiple response, ordering and matching; an unanswered question counts as wrong, and there is no penalty for guessing.
Who should take it — and who should skip it?
Take it if your employer or target employers run on AWS and you want a proctored, no-code AI credential that their engineers will recognise. It suits analysts, project managers, sales engineers, security staff and managers in AWS shops — the wider no-code field is mapped in our no-coding guide.
Skip it if your organisation runs Microsoft — take Azure AI Fundamentals on your own stack instead (its exam is now AI-901, which unlike AIF-C01 expects basic Python; the trade-offs are in our AIF-C01 vs AI-900 comparison) — or if you are an engineer who will be building models and pipelines, in which case go directly to the associate-level exam covered in our AWS ML Engineer Associate study guide. The three-cloud landscape sits in our AWS vs Azure vs Google comparison.
Not sure this is the right one for you?
Tell the picker about your background and what you want the certificate to do, and it narrows the list to the one or two courses we would start with. It suggests only our affiliate partners’ courses, and says so before it suggests anything.
Try the AI Certification Picker →The week-by-week study plan
Built for about five hours a week and ordered by the domain weights above, so the biggest domain gets the most time; compress it if you have more:
- Week 1 — Domain 1, fundamentals of AI and ML (20%): read the exam guide end to end, then work the AI and ML fundamentals in AWS's own training path. Start a running list of every service name you meet and what problem it solves.
- Week 2 — Domain 2, generative-AI fundamentals (24%): the concepts first, then the AWS infrastructure and services the guide names for building generative-AI applications. Take a first practice test at the end of the week to find your weak domains while there is time to fix them.
- Weeks 3–4 — Domain 3, applications of foundation models (28%): design considerations, prompt-engineering techniques, training and fine-tuning, and evaluation, each mapped to the AWS services that deliver it. Turn your service list into scenario flashcards — 'need X, which service?' — because that is the exam's native format.
- Week 5 — Domains 4 and 5, responsible AI and security, compliance and governance (14% each): governance concepts are easy to nod along to and hard to answer precisely, so drill them as scenarios rather than definitions.
- Week 6 — practice and gaps: full-length practice exams under timed conditions, every wrong answer reviewed back to the source material, and the real exam booked once your practice scores sit consistently above the passing line in every domain.
Which resources are actually worth using?
Fewer than the internet suggests. The official exam guide is the syllabus; AWS's own training path covers it in the examiner's vocabulary; and one reputable set of practice exams supplies the drilling (our review of the AWS AI Practitioner practice exams covers the set we would use). A single well-reviewed third-party video course can substitute for the official path if you prefer taught lessons — but adding a second and third course is displacement activity, not preparation. Nothing on this exam requires paid tooling; a free-tier AWS account for poking at the console helps intuition but is optional at this level.
The practice set we mean is Udemy’s AWS Certified AI Practitioner AIF-C01 Practice Exams: a question bank, not a course, with four full practice exams of 260 questions in all and no video, so use it after the study, not instead of it. If you would rather have taught lessons than AWS’s path, the video course we would pick is Udemy’s Ultimate AWS Certified AI Practitioner AIF-C01, about 10 hours with no coding and one practice exam included. Both are one-off Udemy purchases that end in a certificate of completion; the certification itself comes only from passing AWS’s exam.
How should you use practice exams?
As a diagnostic, not a script. Take the first practice test early — after week two — to find your weak domains while there is still time to fix them, not as a graduation ceremony at the end. Review wrong answers to the source material the same day, and track scores by domain rather than overall: a comfortable average can hide a failing domain that the real exam will find. Book the real thing when domain-level scores are consistently above the passing line, and stop drilling the same question bank once you start recognising answers — recognition inflates scores without adding knowledge.
What happens on exam day?
You sit it online proctored or at a test centre, booked through AWS's exam portal. Online proctoring demands a bare desk, a webcam sweep and an interruption-free room — factor that in if home is busy. Tactically: the exam rewards flag-and-return. Scenario questions reward a first fast pass answering everything you are sure of, then a second pass on flagged items; running the clock out on question eleven is the classic self-inflicted fail. Unscored experimental questions may be mixed in, so do not let one bizarre question rattle you.
The table below shows the fee and format of AIF-C01, as AWS publishes them. Every other exam we track is compared on our AI certification exams page.
| Credential | Vendor | Level | Fee | Exam format | Enrol |
|---|---|---|---|---|---|
| AWS Certified AI Practitioner (AIF-C01) | AWS | Foundational | 100 USD | 90 minutes · 65 questions | AWS → |
If you want to pay for the exam before you book it, Udemy sells an AWS Certified AI Practitioner exam voucher; our AI Practitioner review sets out its terms as Udemy states them.
Where most AIF-C01 advice gets it wrong
The exam-dump economy is the big one. Memorising leaked questions violates the exam agreement, risks certificate revocation, and — the practical point — produces certificate holders who cannot answer the first follow-up question at work. The credential's entire value is that it is proctored; cheating it buys a signal you then personally falsify. The subtler error is memorising service names without use-cases. Candidates who can recite the Bedrock feature list still fail scenario questions asking which tool fits a situation — and that mapping, not the vocabulary, is also the part that makes you useful in meetings afterwards.
Verdict
Treat AIF-C01 as a four-to-six-week project: official exam guide, AWS's own training path, one practice-exam set, book when domain scores clear the line. It is the right first proctored credential for anyone in an AWS-stack organisation who wants AI literacy with teeth — and if you are unsure it is the right exam at all, our 2026 rankings and honest take on whether AI certifications are worth it put it in context. Passed it? The technical continuation is the ML Engineer Associate; the broader sequencing lives in our AI certification roadmap. Still weighing options? Our free Picker matches you in two minutes.
Ready to start?
Bought once, with what Udemy calls lifetime access. Udemy's price swings between its list price and a sale price, sometimes within days — check it on the day rather than trusting any figure you read, here or anywhere else.
Frequently asked questions
How long does it take to study for the AWS AI Practitioner?
Around four to six weeks at five hours a week if you do not work in AWS day to day, and two to three weeks if you already live in the console. The variable is scenario fluency rather than concept difficulty — the concepts are foundational, but AWS asks about them in situational form, and that phrasing takes practice to read quickly.
The shape that works inside those weeks matters more than the total. Spend the first fortnight on the official exam guide and the AI, ML and generative-AI fundamentals, keeping a running list of every service name and the problem it solves. Spend the middle weeks turning that list into scenario flashcards — "need X, which service?" — because that is the exam's native format. Reserve the final fortnight for full-length timed practice regardless of your background, and book the real exam once your domain-level scores sit above the line.
What is the passing score for AIF-C01?
700. AWS's exam guide reports results as a scaled score from 100 to 1,000 and sets the minimum passing score at 700, scored compensatorily: you need to pass the exam as a whole, not each domain. Only 50 of the questions count; the other 15 are unscored questions AWS is trialling, and they are not identified. AWS revises its guides, so read the current one before you book rather than a figure from a forum.
The more useful target is the one you set for practice tests. Aim comfortably above the published line in every domain, not just on average: compensatory scoring means one weak domain will not fail you by itself, but a strong overall practice score routinely hides one, and it drags the real total down. Track results per domain, review every wrong answer back to the source material the same day, and stop drilling a question bank once you start recognising the answers — recognition inflates scores without adding knowledge, which is exactly how people walk in over-confident.
Is the AWS AI Practitioner multiple choice?
Mostly. AWS's exam guide lists four question types: multiple choice, multiple response, ordering and matching. No question requires writing or reading code, and there are no hands-on labs at this level. What catches people is not the format but the phrasing: most questions are scenarios ("a company wants to…, which service should it use?"), so you are being tested on mapping situations to services rather than on reciting definitions.
Two tactical points for the sitting itself. The exam rewards flag-and-return: make one fast pass answering everything you are sure of, then a second pass on what you flagged — running the clock out on question eleven is the classic self-inflicted fail. And 15 of the questions are unscored items AWS is trialling, not identified on the exam, so an item that looks bizarre is not evidence you have misjudged the syllabus. Answer it and move on.
Does the AWS AI Practitioner expire?
Yes. AWS AI Practitioner is valid for three years. Before it expires you can recertify by passing the latest version of the exam, or by earning the AWS Certified Machine Learning Engineer – Associate, which recertifies it automatically. AWS sets and occasionally revises those terms, so check the policy that applies when you certify.
This is the structural difference between a vendor exam and a course certificate, and it belongs in the decision rather than in a surprise three years later. Microsoft's fundamentals certifications, including Azure AI Fundamentals, do not expire at all, and the Coursera specializations we review stay on your profile permanently once earned. A vendor credential is a recurring commitment of both money and time — entirely worth it while you work in that ecosystem and the certificate is doing a job, and quietly wasteful once you have moved on.
What should I take after AIF-C01?
It depends on which of the two reasons you took it for. If you are heading toward technical work — building models and pipelines rather than understanding them — the AWS Certified Machine Learning Engineer – Associate (MLA-C01 / MLA-C02) is the real next rung, and it is a different order of commitment: roughly two to three months of preparation, and it does require Python.
If AIF-C01 was your literacy goal, the best next step is not another exam. Spend the following month applying what it taught you to your actual job — one workflow you improve, one thing you build with the services you can now name — because a second foundational badge adds almost nothing a hiring manager can read, while a shipped example answers the question the certificate only raises. Foundational credential, real project, then a deeper exam when the work demands it: that is the order that pays.
Keeping this current. Course formats, prices, and certification exam fees change and vary by region. We review our guides regularly, and we always recommend confirming the specifics on the provider's official page before you enrol.