- What GAF Means on This Site
- Who Issues Generative AI Foundations
- Why the Acronym Causes Confusion
- The Four Exam Domains Behind the Name
- Exam Format and Scoring at a Glance
- Vouchers, Fees, and Delivery Channels
- Who the Credential Is Built For
- Validity and Renewal
- Sequencing Your Preparation by Domain
- Where the Credential Fits in Hiring
- Frequently Asked Questions
- Here, GAF means Generative AI Foundations, a Pearson exam delivered through Certiport within Critical Career Skills.
- The exam objectives are organized into four domains, and no percentage weights are published.
- You get 50 timed minutes, and a scaled score of 700 on a 1-1000 scale is required to pass.
- U.S. Certiport vouchers cost USD 72, or USD 83 with a retake, and work only through Certiport Authorized Testing Centers.
What GAF Means on This Site
If you searched for "GAF meaning," you have probably noticed that three letters can point in several directions. On this site the answer is specific: GAF stands for Generative AI Foundations, an entry-level certification exam about how generative artificial intelligence tools work, how to prompt them well, how to improve their output, and how to use them responsibly.
That definition is the only one this page covers. Other credentials, products, and organizations share the same acronym, and nothing here applies to them. Every fact below describes the Generative AI Foundations exam from Pearson. If you want the shortest possible orientation, our What Is GAF? overview covers the basics, and the GAF Certification page goes into the credential itself.
Who Issues Generative AI Foundations
The exam is issued by Pearson and developed and administered through Certiport, a Pearson VUE business. Within Certiport's catalog it sits under Critical Career Skills (CCS), the family of credentials aimed at workplace readiness skills rather than deep specialist engineering. Pearson announced the certification on October 1, 2024, and the exam objectives document that defines the content carries a 2024 copyright.
One nuance is worth stating plainly: the 2024 copyright identifies the source document for the objectives. It is not a separately verified exam-version designation, so avoid describing the exam as "the 2024 version" in a resume or in conversation. Simply list it as Generative AI Foundations from Pearson/Certiport.
Why the Acronym Causes Confusion
The shorthand is used for several unrelated things, which is why searches like "what does GAF stand for" return a mixed bag. Our companion explainers, What Does GAF Stand For?, What Does GAF Mean?, and What Is A GAF?, address the different ways people phrase the question. The practical rule for exam candidates is simple: if the credential is described as testing generative AI methods, prompt engineering, prompt refinement, and AI ethics, it is the one this site covers. If anything about the description involves a different industry or a different certifying body, you are looking at a different credential, and you should not borrow fees, dates, or pass criteria from it.
| Question | Generative AI Foundations (GAF) answer |
|---|---|
| What does the acronym expand to? | Generative AI Foundations |
| Who issues it? | Pearson, via Certiport (a Pearson VUE business) |
| Which program family? | Critical Career Skills (CCS) |
| Is it vendor-specific? | No, it is brand-agnostic |
| Exam length? | 50 timed minutes |
| Passing score? | 700 on a 1-1000 scaled score |
The Four Exam Domains Behind the Name
The word "Foundations" in the title is earned by the exam's scope. The official objectives are organized into four domains. Certiport's outline lists the domain headings without percentage weights, so any site claiming exact per-domain percentages for this exam is guessing. For a deeper walkthrough, see the GAF Exam Domains guide.
Domain 1: Generative AI Methods and Methodologies
This is the conceptual backbone. Candidates should be able to explain what generative AI is, how it differs from other kinds of software, and how generative systems produce text, images, and other content.
- Core vocabulary: models, training data, inputs, outputs, and generated content
- Where generative tools are strong and where they are unreliable
- How different kinds of generative tools fit different tasks
Domain 2: Basic Prompt Engineering
Prompting is the practical skill the exam rewards most visibly. Expect scenarios where you must recognize what makes a prompt clear, specific, and well-scoped.
- Giving context, a defined task, and constraints in a request
- Specifying audience, tone, format, and length
- Recognizing why a vague prompt yields a vague answer
Domain 3: Prompt Refinement
Writing a first prompt is only half the work. This domain focuses on iterating when the output misses the mark.
- Diagnosing why a response was off-target, incomplete, or inaccurate
- Adjusting wording, adding examples, or narrowing scope to improve results
- Evaluating output critically rather than accepting it at face value
Domain 4: Ethics, Law, and Societal Impact
This domain separates a responsible user from a merely fast one. It covers the human consequences of using generative tools at work and in school.
- Bias, fairness, and the limits of AI-generated content
- Intellectual property, privacy, and data-handling concerns
- Responsible disclosure and accountability when AI assists your work
Exam Format and Scoring at a Glance
The exam is a proctored, computer-based knowledge assessment. Candidates select answers, can flag questions, and can review them before final submission. The timed portion lasts 50 minutes, and that figure excludes the tutorial and administrative seat time, so plan for a somewhat longer visit than the timer alone suggests.
Question count: an honest answer
Sources disagree on how many questions appear. Certiport's tutorial for this exam specifies 40-45 questions, while the current Mindhub voucher page estimates 45-50. Certiport's scoring policy also states that totals differ by exam form. The takeaway: do not memorize a single number. Prepare to work steadily through a form of roughly mid-forties to fifty items and use the flagging feature when you need to move on.
How the score works
The passing mark is 700 on Certiport's 1-1000 scaled-score system. This is not a percentage of questions answered correctly, so you cannot convert 700 into "70 percent right." Scaled scoring accounts for differences between exam forms. Our GAF Passing Score article explains this in more depth, and if you are weighing difficulty, read How Hard Is the GAF Exam?. For anyone looking for historical pass-rate numbers, the honest summary on GAF Pass Rate is that you should be wary of any precise figure that is not published by the issuer.
Key Takeaway
Plan around the scaled score, not a percentage. Aim for consistent, confident performance across all four domains rather than trying to calculate a minimum number of correct answers.
Vouchers, Fees, and Delivery Channels
This is the area where candidates most often make a costly mistake, because two voucher routes exist and they are not interchangeable. For the full cost picture, see GAF Certification Cost.
| Option | Price (U.S.) | Key condition |
|---|---|---|
| Certiport single-exam voucher | USD 72 | Valid only for Certiport Authorized Testing Center (CATC) delivery, in person or remote |
| Certiport voucher with retake | USD 83 | Same CATC-only restriction |
| Pearson Mindhub voucher | USD 85 | Country restrictions apply; must match the country chosen at checkout |
| Mindhub Pro voucher | USD 85 | Web and member prices both listed at USD 85 |
Certiport vouchers may also involve a separate CATC proctoring charge, so ask the center about it before you book. Importantly, the Certiport vouchers cannot be redeemed through OnVUE or at a Pearson VUE Testing Center.
Testing-day rules
- The exam is proctored, and valid identification is required.
- Candidates under 18 need parental consent under CATC rules.
- Books and personal devices must remain inaccessible during the exam.
Who the Credential Is Built For
The exam is described as best suited for candidates age 14 and older, which places it squarely in secondary schools, community programs, workforce-readiness courses, and early-career upskilling. No bachelor's degree is required. The issuer expects about 150 combined hours of instruction and hands-on use of generative AI tools, along with familiarity with productivity applications such as word processors and presentation software.
That 150-hour figure describes candidate preparation. It is not the exam timer and it is not a mandatory course length. If you are checking whether you qualify, the GAF Requirements article walks through eligibility in more detail.
Validity and Renewal
CCS certifications earned on or after June 18, 2025 are valid for five years. To recertify you need a new voucher and must pass the current exam. Earlier nonexpiring CCS certifications retain their existing status. If you earned the credential after the cutoff date, put the renewal window on your calendar rather than assuming it never lapses.
Sequencing Your Preparation by Domain
You do not need an elaborate system for an exam like this, but the order in which you tackle the domains does matter, because the later ones build on the earlier ones. A simple domain-driven sequence works well. For a complete preparation plan, use the GAF Study Guide, and keep the GAF Cheat Sheet handy for last-minute review.
Domain 1: Methods and Methodologies
- Learn the vocabulary and the high-level logic of how generative tools produce content
- Use a generative tool daily so the concepts have real examples attached
Domains 2 and 3: Prompting and Refinement
- Write prompts, deliberately make them worse, then repair them
- Practice diagnosing why an output failed before changing the prompt
Domain 4: Ethics, Law, and Societal Impact
- Work through scenarios involving bias, privacy, and intellectual property
- Take a full timed practice session on the GAF practice test and review every miss
Pairing Domains 2 and 3 in the same week is deliberate: refinement only makes sense once you have written something to refine. Saving ethics for last keeps scenario-style reasoning fresh going into the exam. Our GAF Training page covers course and training options if you prefer structured instruction.
Where the Credential Fits in Hiring
Be realistic about what a foundations credential signals. It demonstrates baseline literacy: you understand what generative AI can and cannot do, you can prompt it competently, and you know the responsible-use issues. That is valuable for students building a profile, for office and administrative staff adopting AI tools, for educators, and for career changers who want a recognized, third-party-issued marker of AI readiness. It is not a substitute for engineering depth, and it should not be presented as one.
Employers who value it tend to be organizations rolling out AI tools broadly and wanting staff who will use them effectively and safely. For a grounded look at roles, see GAF Jobs. For earnings context, the GAF Salary Guide explains why we avoid quoting a single dollar figure, and Is the GAF Certification Worth It? weighs the investment against likely benefits. If you want a plain-language definition of the credential itself, What Is GAF Certification? is the place to start.
Key Takeaway
Treat Generative AI Foundations as proof of responsible, practical AI literacy. Pair it with a portfolio of real work samples, such as well-crafted prompts and documented before-and-after refinements, to make the credential more persuasive.
Frequently Asked Questions
GAF means Generative AI Foundations, a Pearson certification exam delivered through Certiport within Critical Career Skills. It tests generative AI methods, basic prompt engineering, prompt refinement, and ethics, law, and societal impact.
The timed portion is 50 minutes, excluding the tutorial and administrative time. You need a scaled score of 700 on Certiport's 1-1000 scale, which is not a percentage of questions answered correctly.
Sources differ. Certiport's tutorial specifies 40-45 questions, while the Mindhub voucher page estimates 45-50, and totals vary by exam form. Prepare for a form in that general range.
In the U.S., a Certiport voucher is USD 72, or USD 83 with a retake, plus any CATC proctoring charge, and it works only at Certiport Authorized Testing Centers. A Mindhub voucher is USD 85 with country restrictions. Match your voucher to your testing channel before paying.
CCS certifications earned on or after June 18, 2025 are valid for five years, and recertification requires a new voucher and passing the current exam. Earlier nonexpiring certifications keep their existing status.
Understanding what the acronym stands for is the first step. The next is testing yourself against the four domains, and you can start by working through questions on the GAF Exam Prep practice site.