- ITS-AI here means the Pearson/Certiport IT Specialist - Artificial Intelligence certification, a six-domain, proctored knowledge exam.
- The exam lists 45-50 questions with a 50-minute maximum; the passing score is 700 on a 1-1000 scale.
- The English exam content update took effect June 30, 2026, adding feature engineering, NLP, explainability and CI/CD coverage.
- Awards earned on or after June 18, 2025 are valid for five years; recertification means passing the current exam.
What ITS-AI Actually Is
On this site, ITS-AI refers to one credential only: the Information Technology Specialist - Artificial Intelligence Certification, published in the public catalogue as IT Specialist - Artificial Intelligence. It is a computer-based knowledge assessment that verifies whether a candidate understands the full lifecycle of a practical AI project, from defining the problem through monitoring a model after it ships.
The acronym is shared by other, unrelated credentials in the wider market, so it is worth being precise. Everything on this page concerns the Pearson-issued IT Specialist exam, catalogued by Pearson under identifier INF-307. If you have seen exam fees, domain weights or salary claims attached to a different "ITS-AI," those figures do not transfer here. For a shorter orientation to the name itself, see our explainers on what ITS-AI stands for and the meaning of ITS-AI.
The exam is aimed at people building foundational, job-relevant AI skills rather than researchers. It tests whether you can reason about the choices a practitioner makes: is this problem suited to AI at all, is the data usable, which kind of model fits, how will the application be secured and deployed, and what happens when performance degrades in production.
Who Issues and Administers It
The credential is issued by Pearson and administered through Certiport, a Pearson VUE business. Certiport runs the IT Specialist program, which covers a family of entry-level technology certifications, and ITS-AI is one of them. Candidates take the exam either through the Pearson VUE channel (test centers or OnVUE online proctoring) or through the separate Certiport Authorized Testing Center (CATC) channel, which is common in schools and training organizations. The differences between those channels matter for pricing, and we cover them below.
The Six Exam Objectives
The current objective document is titled IT Specialist Exam Objectives - Artificial Intelligence (copyright 2026). It lists six domains. Two points are worth stating clearly. First, the six headings are official exam objectives, not a preparation curriculum, so they tell you what is measured rather than how to learn it. Second, Pearson does not publish percentage weights for these domains, so any site quoting exact weightings is guessing. For a deeper walkthrough of every content area, read our complete guide to all six ITS-AI content areas.
Domain 1: AI Problem Definition
Before any code, you must decide whether a problem should be solved with AI and how success will be judged.
- Framing a business need as a machine learning task
- Feasibility: whether data, resources and expected value justify an AI approach
- Choosing evaluation criteria that match the goal
- Considering human interaction with the system
Domain 2: Data Collection and Transformation
Models are only as dependable as the data behind them, and this domain covers getting that data ready.
- Sourcing, collecting and assessing data quality
- Cleaning, transforming and preparing datasets
- Recognizing bias, gaps and leakage risks in data
Domain 3: Feature Engineering
The revised exam treats feature engineering as its own domain rather than folding it into data preparation.
- Selecting, creating and transforming features
- Understanding how feature choices affect model behavior
- Preparing inputs for different model types
Domain 4: AI Algorithms and Models
The conceptual core: matching techniques to tasks and evaluating the results.
- Selecting appropriate algorithms and model families
- Natural language processing concepts
- Model-evaluation metrics and interpreting them correctly
- Explainability: why a model produced a given output
Domain 5: Application Integrity and Deployment
This is the exact wording used in the revised objective document. It covers turning a model into a trustworthy application.
- Application integration (objective 5.1)
- CI/CD practices for AI applications
- Security considerations for AI systems
- Deployment decisions and release practices
Domain 6: Maintaining and Monitoring AI in Production
Shipping a model is not the end of the project, and the final domain reflects that.
- Monitoring performance after deployment
- Detecting and responding to drift
- Planning for retraining and decommissioning (the revised document ends at objective 6.5)
Current vs. Legacy Outlines: A Source Conflict Worth Knowing
Candidates frequently study from the wrong outline, so this deserves its own section. Pearson's general IT Specialist resources page still links an older, copyright-2025 outline with five domains. The revised copyright-2026 document, retrievable from Certiport's exam page, has six. The older outline should not be treated as complete current English coverage.
The practical differences are substantive. The current coverage separates out feature engineering, natural language processing, explainability, human interaction, feasibility assessment, model-evaluation metrics, CI/CD, security, drift and decommissioning. Certiport announced the update on April 29, 2026, and listed the English content update as taking effect on June 30, 2026 in its exam content updates archive. Rollout dates for other languages differ, so non-English candidates should confirm their own date.
Format, Length and Scoring
The exam is a computer-based, proctored knowledge assessment. Pearson's MindhubPro listing estimates 45-50 questions with a 50-minute maximum. Treat the question count as a range rather than a fixed form size: the number you see on test day can vary. The 50 minutes refers to timed exam time and does not include administrative steps or the on-screen tutorial.
Certiport's older generic tutorial (copyright 2021) mentions a 36-40 question estimate for IT Specialist exams broadly. That figure predates the exam-specific listing, so rely on the 45-50 range for ITS-AI. The tutorial is still useful for one thing: it demonstrates how answer options are navigated, which is worth a quick look so the interface is not a surprise.
| Exam Detail | ITS-AI Value |
|---|---|
| Issuer / administrator | Pearson; Certiport (a Pearson VUE business) |
| Questions | Estimated 45-50 (variable range) |
| Time limit | 50 minutes maximum, timed exam time only |
| Passing score | 700 on a 1-1000 scaled score |
| Domains | Six, with no official percentage weights published |
| Delivery | Pearson VUE centers, OnVUE, or CATC |
Understanding the 700 passing score
The passing score is 700 on a 1-1000 scale. This is a scaled score, not a raw 70% requirement, so you cannot convert it directly into a number of questions answered correctly. Certiport does not publish pass rates, which means any quoted pass percentage for this exam is unsupported. Our pages on the ITS-AI passing score and what pass-rate data does and does not show go into more detail.
Fees, Vouchers and Delivery Channels
Pricing depends on which channel you use, and the two channels are not interchangeable.
| Option | Price | Key Conditions |
|---|---|---|
| Exam-only voucher (MindhubPro) | USD 127 web/member pricing | Country-restricted, unavailable in listed markets, expires 12 months after purchase |
| CATC voucher with retake | USD 133 | US-only; proctoring may cost extra; not usable at Pearson VUE centers or through OnVUE |
Two details trip people up. First, the 12-month voucher expiry is separate from credential validity: an unused voucher lapses a year after purchase regardless of how long a passed credential would last. Second, the CATC retake voucher only works inside the CATC channel, so do not buy it if you plan to test at a Pearson VUE center or from home. For the full cost picture, including what might be extra, see our ITS-AI certification cost breakdown.
Test-day mechanics
If you test through OnVUE, expect a required system check and a live webcam proctor. At a CATC, conditions include photo and signature ID, a non-disclosure agreement, parental consent for candidates under 18, and no personal paper or communication devices in the testing area. Scheduling specifics are covered in our guide to ITS-AI exam dates and scheduling.
Readiness and Prerequisites
There is no degree requirement; the stated expectation is knowledge of the objective domains. Certiport's program materials suggest roughly 150 hours of preparation, and the objective profiles describe candidates with at least 150 hours of experience or instruction. Read these as readiness profiles, not as a verified application requirement that you log hours before registering. Nobody checks a timesheet, but the figure is a useful sanity check on how much ground the six domains cover. The details are collected in our ITS-AI requirements guide.
Key Takeaway
Treat 150 hours as a planning benchmark, not a gate. If you already work with data pipelines or model evaluation, you may need less time in Domains 2 and 4 and more in Domains 5 and 6, where CI/CD, security and drift are often less familiar to newcomers.
How Long the Credential Lasts
Awards earned on or after June 18, 2025 are valid for five years. Awards earned before that date remain nonexpiring. When a five-year credential expires, recertification means buying a new voucher and passing the current exam, which matters in a field where objectives change as quickly as AI does. Given that the exam content was substantially revised in 2026, a future recertifier should expect a different blueprint from today's.
Who Benefits From It
ITS-AI is an entry-level, foundational credential, so it works best as evidence of structured knowledge rather than as a substitute for experience. It tends to suit several groups:
- Students and career changers who need a recognized, vendor-backed signal that they understand the AI project lifecycle.
- IT support, analyst and junior developer staff moving toward AI-adjacent roles in their organizations.
- Educators and training programs that use the IT Specialist family through CATCs to give learners a measurable outcome.
We deliberately do not quote salary figures here, because none are published for this specific credential. If you are weighing the investment, our analyses of ITS-AI jobs, the salary picture and whether the certification is worth it address the question without inventing numbers. Broader context lives on our ITS-AI certification overview.
Sequencing Your Preparation Around the Six Domains
You do not need an elaborate method, but order matters because the domains build on each other. Here is a domain-driven sequence that mirrors the project lifecycle the exam follows. Adjust the pacing to your own background and total hours.
Problem framing and data
- Practice judging feasibility and success criteria (Domain 1)
- Work through data quality, cleaning and bias scenarios (Domain 2)
Features and models
- Study feature creation and transformation (Domain 3)
- Learn when each algorithm family fits, plus NLP, evaluation metrics and explainability (Domain 4)
Deployment and operations
- Cover integration, CI/CD and security (Domain 5)
- Cover drift, monitoring and decommissioning (Domain 6)
Timed practice
- Take full-length practice sets under a 50-minute limit
- Review misses by domain, not by question
Domains 5 and 6 deserve extra attention because they are the newest and most operational parts of the blueprint, and candidates who study only modeling often underestimate them. For a fuller plan, see our ITS-AI study guide, and for a compact last-minute review, the ITS-AI cheat sheet. When you are ready to test yourself against exam-style questions, try the ITS-AI practice tests, and revisit the difficulty guide to calibrate expectations.
Frequently Asked Questions
It stands for the Information Technology Specialist - Artificial Intelligence Certification, issued by Pearson and administered by Certiport. The acronym is shared by other credentials, but none of their details apply here.
The exam is estimated at 45-50 questions with a 50-minute maximum. The question count is a variable range, and the time limit covers timed exam time only, not administration or tutorials.
You need 700 on a scaled range of 1-1000. It is a scaled score rather than a raw 70% requirement, and Certiport does not publish pass rates.
Pearson MindhubPro displays a USD 127 exam-only voucher that is country-restricted and expires 12 months after purchase. A separate US-only CATC voucher with retake is USD 133, and proctoring may cost extra.
Awards earned on or after June 18, 2025 are valid for five years, and earlier awards do not expire. Recertification requires a new voucher and passing the current exam.