- What the IT Specialist - Artificial Intelligence Certification Actually Is
- The Revised Six-Domain Outline and Why the Version Matters
- What Each Domain Expects You to Know
- Exam Format, Timing and Scoring
- Registration Channels, Vouchers and Fees
- Readiness: The 150-Hour Profile and Who Should Sit the Exam
- Sequencing the Six Domains Across Your Preparation
- Where the Credential Fits in a Career
- Validity, Expiry and Recertification
- Frequently Asked Questions
- The credential is IT Specialist - Artificial Intelligence, issued by Pearson and administered through Certiport, a Pearson VUE business.
- The current English outline has six domains, updated June 30, 2026; Pearson's legacy five-domain PDF is outdated.
- The passing score is 700 on a 1-1000 scaled range, not a raw 70 percent.
- Expect an estimated 45-50 questions with a 50-minute timed maximum.
What the IT Specialist - Artificial Intelligence Certification Actually Is
The ITS-AI Certification is the exam-and-credential pairing officially titled IT Specialist - Artificial Intelligence. It is issued by Pearson and delivered through Certiport, which operates as a Pearson VUE business. Pearson's catalogue tracks it under the identifier INF-307, and you will see that "307" surface in objective documents and in some retail practice-bank product names.
It is a computer-based, proctored knowledge assessment. That wording matters: this is not a lab-based or project-based credential. You demonstrate what you know about the AI lifecycle by answering questions under timed, supervised conditions. There is no degree requirement; the only gate is your command of the published exam objectives.
If you are brand new to the name and want the plain-language basics first, our explainers on what ITS-AI certification is and what ITS-AI stands for cover the terminology. This article goes further into structure, mechanics and preparation planning for this specific credential.
The Revised Six-Domain Outline and Why the Version Matters
The single most important thing to verify before you study is which outline you are using. Certiport's exam content updates archive lists an English update dated June 30, 2026, preceded by an announcement on April 29, 2026 titled "We're updating the IT Specialist AI certification." The revised objectives document (copyright 2026) organizes the exam into six domains. Rollout dates for other languages differ, so if you are testing in a language other than English, confirm your own timeline.
Several topics that were folded together or absent in older material now stand on their own in the current outline: feature engineering, natural language processing, explainability, human interaction, feasibility assessment, model-evaluation metrics, CI/CD, security, drift and decommissioning. Some retail practice products still carry legacy headings, and their per-domain item counts are not official exam weights. Certiport publishes no official percentage weights for these domains, so any source claiming exact domain percentages is not drawing on the issuer.
For a full tour of the structure, see our complete guide to all six ITS-AI content areas.
What Each Domain Expects You to Know
The six domain names below are official exam objective headings, not a preparation curriculum. They describe what is measured, not how you should learn it. Use them as a checklist against your study materials.
Domain 1: AI Problem Definition
This domain is about deciding whether and how AI should be applied before any model is built. Expect scenario thinking: given a business need, is AI appropriate, and is the project feasible?
- Framing a business problem as an AI task rather than defaulting to a model
- Feasibility assessment: data availability, constraints, and expected value
- Human interaction considerations in how people will use or be affected by the system
Domain 2: Data Collection and Transformation
Models are only as good as the data feeding them. This domain covers acquiring, cleaning and reshaping data into a usable state.
- Sourcing and assembling datasets from different origins
- Cleaning, handling missing values, and correcting inconsistencies
- Transforming data into formats suitable for downstream modelling
Domain 3: Feature Engineering
Feature engineering is now its own domain rather than a footnote to data prep. Know how raw inputs become the signals a model learns from.
- Selecting, creating and transforming features
- Understanding how feature choices affect model behavior and performance
- Recognizing how text-oriented work such as natural language processing handles features differently from tabular data
Domain 4: AI Algorithms and Models
This is the conceptual core: what kinds of models exist, what they suit, and how to judge them.
- Matching algorithm families to problem types
- Model-evaluation metrics: knowing what each metric measures and when it misleads
- Explainability: why a model produced an output and how to communicate it
- Natural language processing concepts at the level the objectives describe
Domain 5: Application Integrity and Deployment
This domain name preserves the revised PDF's exact wording. Application integration remains objective 5.1, so do not skip it assuming it moved elsewhere.
- Integrating a trained model into an application
- CI/CD concepts as they apply to AI delivery
- Security considerations for AI systems and the data they handle
Domain 6: Maintaining and Monitoring AI in Production
The lifecycle does not end at launch. The revised outline runs through to objective 6.5, covering what happens after deployment.
- Monitoring performance after release
- Detecting and responding to drift
- Decommissioning systems responsibly when they reach end of life
Exam Format, Timing and Scoring
Question count and time limit
Pearson's MindhubPro listing estimates 45-50 questions with a 50-minute maximum. Treat the question count as a variable range, not a fixed form size. The 50 minutes is timed exam time and excludes administration and tutorial screens, per Certiport's exam-lengths page. Certiport's older general IT Specialist tutorial document cites a generic 36-40-question estimate; it predates the exam-specific listing and should not override it. That same tutorial remains a useful reference for how answer-option navigation works on screen.
Scoring
The passing score is 700 on a 1-1000 scale. This is a scaled score, so do not translate it into "I need 70 percent of questions right." Scaling accounts for differences between exam forms, and the relationship between raw correct answers and your reported score is not published as a simple percentage. Certiport does not publish pass rates, so any specific pass-rate figure you encounter is unverified; our piece on what the pass-rate data does and does not show and the dedicated explainer on the ITS-AI passing score unpack this further.
| Item | What is documented |
|---|---|
| Issuer / administrator | Pearson; Certiport, a Pearson VUE business |
| Delivery | Computer-based, proctored |
| Estimated questions | 45-50 (variable range) |
| Time limit | 50 minutes maximum, timed exam time only |
| Passing score | 700 on a 1-1000 scaled range |
| Official domain weights | Not published |
| Pass rate | Not published |
With roughly one minute per question available, the format rewards clean recall and quick scenario recognition more than long deliberation. If you are weighing how demanding that is, our difficulty guide discusses the experience in context.
Registration Channels, Vouchers and Fees
There are two distinct routes to sit the exam, and they are not interchangeable. Mixing them up is one of the most avoidable mistakes candidates make.
Route 1: Pearson VUE test centers or OnVUE
You can test at a Pearson VUE center or remotely through OnVUE. OnVUE requires a system check and a live webcam proctor. Pearson MindhubPro displays an exam-only price of USD 127 for web/member pricing. The voucher is country-restricted and unavailable in certain listed markets, so check availability for your country before purchasing. The voucher also expires 12 months after purchase. That is a voucher deadline and is entirely separate from how long the credential itself stays valid once earned.
Route 2: Certiport Authorized Testing Center (CATC)
The CATC channel is a separate path, often used through schools and training providers. A US-only CATC voucher that includes a retake is listed at USD 133, and proctoring may cost extra. Critically, that voucher is not usable at Pearson VUE centers or through OnVUE. CATC testing conditions include photo and signature ID, a signed NDA, parental consent for candidates under 18, and a rule against personal paper or communication devices in the testing area.
| Factor | Pearson VUE / OnVUE route | CATC route |
|---|---|---|
| Listed price | USD 127 (web/member pricing) | USD 133, US-only, includes retake |
| Retake included | Not stated for the exam-only voucher | Yes |
| Voucher usable at the other channel | Exam-only voucher is for this route | No; not usable at VUE centers or OnVUE |
| Extra proctoring cost | Not stated | May apply |
| Voucher expiry | 12 months after purchase | Check the listing at purchase |
Readiness: The 150-Hour Profile and Who Should Sit the Exam
The program suggests roughly 150 hours of preparation, and the objective profiles describe candidates with at least 150 hours of relevant experience or study. Treat this as a readiness profile, not a verified requirement: there is no logged-hours application step, and nobody checks a timesheet before you register. It is guidance about the depth of knowledge the exam assumes.
Beyond that, there is no degree requirement. The credential is positioned for people building foundational, vendor-neutral understanding of the AI lifecycle: students, career changers, IT staff expanding into AI-adjacent work, and early-career technologists. If you want the eligibility picture spelled out, read our guide to ITS-AI requirements and prerequisites.
Key Takeaway
The 150-hour figure is a calibration tool, not a gate. If you can already explain every objective from 1.1 through 6.5 without notes, you are likely past the profile regardless of hours logged; if you cannot, more hours alone will not fix gaps in specific domains.
Sequencing the Six Domains Across Your Preparation
Because the domains follow the lifecycle of an AI project, the most natural study order mirrors it, with deliberate overlap where topics depend on each other. Here is one way to distribute a roughly 150-hour effort across six weeks of focused work, adjusted for the dependencies in this particular exam.
Domain 1: AI Problem Definition
- Practice judging feasibility and framing problems before choosing methods
- Include human interaction considerations early; they thread through later domains
Domain 2: Data Collection and Transformation
- Work through cleaning, missing data and format transformation scenarios
- Note which data problems later constrain feature and model choices
Domain 3: Feature Engineering
- Study it now, while data handling is fresh, since it builds directly on Domain 2
- Compare structured-data features with text-oriented inputs
Domain 4: AI Algorithms and Models
- Give this the most time; it carries metrics, explainability and NLP concepts
- Drill when each evaluation metric is appropriate and when it misleads
Domains 5 and 6: Deployment and Production
- Cover integration (5.1), CI/CD and security together as one delivery story
- Pair drift, monitoring and decommissioning with the deployment material
Integration and timed practice
- Run full timed sets against the 50-minute ceiling
- Revisit weak objectives and confirm your materials reflect the six-domain outline
Because official weights are unpublished, do not trust a plan that assigns precise percentages to each domain. Spread effort sensibly, and use timed practice questions to find where you actually lose points rather than guessing. A broader walkthrough is available in our ITS-AI study guide, and the one-page cheat sheet works well for final review.
Where the Credential Fits in a Career
This is an entry-level, foundational credential. It signals that you understand the end-to-end lifecycle of an AI project, from deciding whether AI is the right tool through to retiring a system, rather than deep specialization in a single framework or cloud platform. That breadth makes it most useful as a first AI credential or as a bridge for people moving from general IT, data or development roles toward AI-adjacent responsibilities.
Employers who value it tend to be those looking for staff who can communicate across the lifecycle: understanding why data quality, explainability, security and monitoring matter, not just how to train a model. Roles that commonly reference this kind of foundation include junior AI and data support positions, IT generalists taking on AI projects, and technical staff in organizations adopting AI tools. We avoid quoting salary numbers here because no verified figure is tied to this specific credential; for a qualitative look at the market, see our pieces on ITS-AI jobs, the salary analysis, and the return-on-investment discussion.
Validity, Expiry and Recertification
Credential validity depends on when you earned it. Awards earned on or after 2025-06-18 are valid for five years. Awards earned before that date remain nonexpiring. To recertify an expiring credential, you purchase a new voucher and pass the current version of the exam. Because the exam content was revised in 2026, a recertification attempt means studying the current six-domain outline rather than whatever you learned originally.
Keep the two clocks distinct: the voucher expires 12 months after purchase, while the credential itself, once earned, follows the five-year rule above. Confusing them can lead you to rush an exam unnecessarily or to assume a credential has lapsed when it has not.
Frequently Asked Questions
Pearson issues it, and Certiport, a Pearson VUE business, administers the exam. It is listed in Pearson's catalogue under identifier INF-307.
Six: AI Problem Definition; Data Collection and Transformation; Feature Engineering; AI Algorithms and Models; Application Integrity and Deployment; and Maintaining and Monitoring AI in Production. The English update took effect June 30, 2026. An older five-domain outline still circulates and should not be used as complete current coverage.
You need 700 on a 1-1000 scaled range. It is a scaled score, not a raw 70 percent, and Certiport does not publish pass rates.
No. The US-only CATC voucher listed at USD 133 works only through the Certiport Authorized Testing Center channel, and proctoring may cost extra. The exam-only voucher at USD 127 is the route for Pearson VUE centers and OnVUE, subject to country availability.
No. The program suggests approximately 150 hours and objective profiles describe at least that much, but this is a readiness profile rather than a verified application requirement. There is no degree requirement beyond knowing the objective domains.