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.
A plain guide to AI+ Security Practitioner (AT-2101) as of October 2026. It covers the exam, the twelve modules and their weights, the exam and renewal fees, what AI CERTs says about accreditation, and who the certification actually suits.
AI+ Security Practitioner is AI CERTs’ foundational certification for people starting out in AI-enabled cybersecurity, tested on a single 50-question exam, AT-2101.
AI CERTs says it validates “foundational knowledge of AI-driven cybersecurity concepts” along with security principles, threats and controls. The blueprint backs that up. Modules 1 through 5, worth 41 percent of the exam, are plain security and IT groundwork: Linux, networking, Python, common threats, cryptography and identity. Only then does it move into machine learning for detection and threat hunting, LLM security, AI red teaming concepts, incident response and governance. A capstone module ties it together, and on the exam those capstone items are scenario-based multiple-choice questions, not a project you hand in.
One naming note before you go searching. AI CERTs’ training site and store list AT-2101 as AI+ Security Practitioner, “formerly known as AI+ Security Level 1.” That older name is still the one on the certification body site, in the certification scheme and on Training Camp’s course page, and the exam code is the same under both. It has no mandatory prerequisites and is issued by AI CERTs, a DBA of Sarder Inc. of New York.
Modules 1 to 5. Linux, networking, Python, threats and cryptography. 41% of the exam.
Modules 6 and 7. Machine learning basics and AI for threat hunting. 18% of the exam.
Modules 8 and 9. LLM security, the OWASP LLM risks and AI red teaming. 17% of the exam.
Modules 10 to 12. SOC work, governance and a capstone scenario. 24% of the exam.
Four things worth knowing before you line it up against the CompTIA or ISACA AI credentials.
CompTIA SecAI+ recommends three to four years in IT, and ISACA AAISM requires an active CISM or CISSP. AT-2101 assumes neither. Modules 1 through 5 cover Linux, networking, Python, threats and cryptography before any machine learning shows up. The AI half then works through AI-driven detection, LLM security and the OWASP LLM risks, and AI red teaming concepts. No single module dominates, either. Seven of the twelve sit at 9 percent each.
AI CERTs certifications are valid for one year. The only way to renew is a proctored multiple-choice re-exam before the certificate expires, and the August 2026 Candidate Handbook lists that fee at $95. There are no continuing education credits to bank, so budget for a yearly exam rather than a three-year cycle.
AI CERTs’ next two levels are AI+ Security Expert (AT-2102) and AI+ Security Strategist (AT-2103). Each level is a separate scheme with its own exam. Our map of every major AI certification in 2026 shows where the AI CERTs track sits next to the CompTIA, ISACA and IAPP options.
AI CERTs states that it operates as an independent certification body under ISO/IEC 17024:2012. On ANAB, its accreditation page says only that it is “actively pursuing ANAB Accreditation.” So AI CERTs does not claim ANAB accreditation for AT-2101 today.
There is no DoD 8140 claim for it either. Our explainer on what makes a certification DoD 8140 approved covers why accreditation is one of the two tests a credential has to pass to reach the matrix.
The certification, the AT-2101 exam and the renewal rules as AI CERTs publishes them in October 2026.
AI+ Security Practitioner is a starting credential. Where you go next depends on whether you stay inside the AI CERTs track, add a core security cert that DoD 8140 recognizes, or move up to a mid-level AI security exam once you have a few years in the field.
AI CERTs’ next two security levels are AI+ Security Expert (AT-2102) and AI+ Security Strategist (AT-2103). The AT-2101 scheme is clear that there is no automatic upgrade path. Each level is its own scheme, so you apply for and pass each exam separately.
If a job posting or a federal contract names DoD 8140, Security+ is the cert to add. It is approved for multiple DCWF work roles, runs up to 90 questions in 90 minutes and stays valid for three years. SY0-801 replaces SY0-701 starting November 17, 2026.
Once you have a few years in security, SecAI+ (CY0-001) is CompTIA’s mid-level AI security exam. CompTIA recommends three to four years of IT experience with at least two in cybersecurity, and Securing AI Systems alone is 40 percent of the exam.
It sits near the start. Some groundwork helps first, then the path splits three ways: deeper into AI CERTs, into the core security certs, or up to a mid-level AI security exam.
Helpful, not required
The foundational AI security credential
Pick the direction that fits the job
Two questions to settle before you buy a voucher: can you sit it, and should this be the AI security cert you pick. Straight answers to both.
No mandatory prerequisites.
The AT-2101 scheme sets no education, experience or prior-certification requirement. You buy the exam from AI CERTs, accept the candidate agreement and test online through Proctor 365 with a government-issued ID. There is no test center option. Plan on a webcam, a single monitor, a quiet room and at least 1.5 Mbps of upload speed.
AI CERTs lists recommended knowledge.
AI CERTs suggests a basic grasp of operating systems, networking, introductory AI and security concepts, and some programming, with Python preferred. None of it is required. Python for Security and Automation is still 9 percent of the exam, though, so if you have never written a loop or a function, give Python basics a week before you start.
AI CERTs maps AT-2101 to seven job roles. These six are the ones a foundational credential can realistically support early on, usually alongside hands-on experience or a core security cert.
Watches alerts, triages events and escalates incidents in a security operations center. Modules 7 and 10 (AI-assisted SOC operations, SIEM concepts, alert management) line up with the day-to-day work.
Monitors security events, analyzes threats and protects enterprise networks. The foundation modules on networking, cryptography and identity carry most of the weight in this role.
Finds weaknesses in AI applications, watches for AI-specific threats and applies controls. Module 8, with its focus on LLM security and the OWASP LLM risks, is the center of this job.
Looks for attackers who slipped past automated defenses. Module 7 covers threat hunting with MITRE ATT&CK mapping, IOC analysis and anomaly detection, which is the toolkit a junior hunter starts with.
Investigates incidents, contains threats and coordinates recovery. Module 10 walks through the incident response lifecycle, evidence handling and malware analysis.
Assesses AI-related security risk and helps build governance for responsible AI use. Module 11 on governance, compliance and privacy is the bridge into this work.
All three test security with some AI in the mix, but they sit at different levels and renew very differently. AI+ Security Practitioner is an AI-first starter, Security+ is the established security baseline, and SecAI+ is the AI security exam for people with a few years behind them.
| AI+ Security Practitioner | CompTIA SecAI+ | CompTIA Security+ | |
|---|---|---|---|
| Issuer | AI CERTs | CompTIA | CompTIA |
| Exam Code | AT-2101 | CY0-001 | SY0-701 (SY0-801 from Nov 17, 2026) |
| Level | Foundational | Specialty (mid-level) | Entry-level core security |
| Questions and Time | 50 questions, 90 minutes | Up to 60 questions, 60 minutes | Up to 90 questions, 90 minutes |
| Question Types | Multiple choice only | Multiple choice and performance-based | Multiple choice and performance-based |
| Passing Score | 70% (35 of 50) | 600 (100 to 900 scale) | 750 (100 to 900 scale) |
| Exam Cost | $195 exam only | $298 | $439 |
| Delivery | Online, AI-proctored only | Pearson VUE or OnVUE online | Pearson VUE or online proctoring |
| Prerequisites | None (OS, networking, Python basics rec.) | None (3 to 4 years IT rec.) | None (Network+ and 2 years IT rec.) |
| Validity | 1 year, renewed by re-exam | 3 years, continuing education | 3 years, 50 CEUs |
| Accreditation | ANAB being pursued, per AI CERTs | ANAB (ISO/IEC 17024) | ANAB (ISO/IEC 17024) |
| DoD 8140 | No claim | Not listed (Matrix V2.1) | Approved for multiple DCWF roles |
| Best For | Early-career staff adding AI to security basics | Experienced practitioners securing AI systems | A recognized baseline and DoD 8140 roles |
The validity row is the one people skip past. A one-year cycle means a $95 re-exam every year to stay current, against three years for either CompTIA cert. Pick by where you are now and whether a DoD 8140 requirement is part of the job.
Certification Matchups: the published facts, strengths and weak spots of 33 certifications, side by side.
Our AT-2101 boot camp runs five days, live online or in person, on official AI CERTs courseware with videos, e-books and quizzes. Hands-on labs and a capstone project run in a provided lab environment, so there is nothing to install. Your AI CERTs exam voucher and a retake are included, along with a mobile learning app, exam-style practice questions and 180 days of one-on-one coaching after class. You sit AT-2101 online through AI CERTs’ remote proctoring.
Where AI+ Security Practitioner sits among the 2026 AI certifications, AI CERTs’ ethical hacking credential, the OWASP agentic risks, and how to study from an exam blueprint.
A cert-by-cert rundown of the 2026 AI credentials, AI CERTs’ AI+ Security among them, with who each one is built for and where it wins or loses against the alternatives.
What AI CERTs’ AI+ Ethical Hacker covers, from AI-assisted reconnaissance to machine learning for finding weaknesses, and why it adds to classic ethical hacking rather than replacing it.
The ten agentic AI risks, ASI01 through ASI10, from the OWASP GenAI Security Project, and why agents that call tools fail differently from chatbots.
A straight look at whether SecAI+ pays off yet, written by someone who sat the beta, and who should hold off for now.
The two tests a certification has to pass to reach the DoD 8140 matrix, ANAB accreditation under ISO/IEC 17024 and a 70 percent work role match.
How to read an exam blueprint, what the domain weights and verbs tell you, and how to turn the document into a study plan instead of skimming it once.
Why CySA+ is the usual next step after Security+, when it isn’t, and how to pick the depth exam that matches the job you want.
The AT-2101 blueprint spreads the exam across twelve modules, with none above 9 percent. At 50 questions, a 9 percent module works out to about four or five questions, so there is no single area big enough to carry you. Click any module for what it covers.
How hardware, software, operating systems and Linux fit together. Linux commands, file systems, user management, permissions, authentication concepts and access control mechanisms, and how they support security operations.
IP and MAC addressing, ports, protocols, DNS and TCP/IP communication, plus how firewalls, IDS/IPS, VPNs and network monitoring help protect an environment and spot suspicious traffic.
Python basics (variables, conditions, loops, functions, modules, file handling and error handling) put to work on log analysis, automation scripts, data processing and AI-driven security workflows.
The CIA triad, threats, vulnerabilities, risk, attack surfaces, defense in depth and security controls, with malware, phishing, ransomware, social engineering and the common industry security frameworks.
Encryption, hashing, digital signatures and TLS, then authentication methods, identity management, access control, password security, multi-factor authentication and the usual identity-related risks.
AI, machine learning and deep learning basics. Supervised, unsupervised and reinforcement learning, the machine learning lifecycle, datasets, features, labels, model evaluation and where AI shows up in security tools.
Behavioral analytics, anomaly detection, machine learning based detection, detection engineering and threat intelligence, plus threat hunting, MITRE ATT&CK mapping, IOC analysis and AI-assisted SOC operations.
Large language models, generative AI, AI copilots and retrieval-augmented generation. The OWASP LLM risks (prompt injection, sensitive information disclosure, insecure outputs, excessive agency), responsible AI and human oversight.
AI threat modeling, attack surfaces and the STRIDE methodology, adversarial AI techniques and attacks on AI systems, and the AI red teaming and security testing approaches used to harden them.
SOC operations, SIEM concepts, alert management, the incident response lifecycle, detection workflows and evidence handling, plus malware analysis, threat investigation and where AI helps the SOC.
Security governance, risk management, AI governance, compliance requirements, privacy principles and responsible AI practice, including how organizations set policy, manage AI risk and protect sensitive data.
An end-to-end AI security assessment in five phases: reconnaissance and environment review, detection and threat analysis, AI security assessment, incident response, and governance and reporting. On the exam these items arrive as scenario-based multiple-choice questions.
Modules and weights are the AT-2101 exam blueprint on AI CERTs’ AI+ Security Practitioner page and store, checked October 21, 2026. The certification scheme on ai-certs.org (exam code AIC-SEC-101, V3.0 revised March 18, 2026) still shows an older 11-domain outline that runs from Introduction to Cybersecurity at 6 percent to a Case Study domain at 10 percent. Ask AI CERTs which outline your exam form follows before you book.
The questions candidates ask most often when they look into the AI CERTs AI+ Security Practitioner certification.
AI+ Security Practitioner is a foundational certification from AI CERTs that covers cybersecurity basics and how AI is used to detect, investigate and defend against threats. It is tested on one exam, AT-2101, with 50 multiple-choice questions. AI CERTs aims it at early-career professionals, IT staff moving into security, and students, and says it is not meant to test advanced or expert-level skills.
No. It is the same AT-2101 exam under a new name. As of October 2026, AI CERTs’ training site and store list it as AI+ Security Practitioner and note that it was formerly the first of AI CERTs’ three numbered security levels. The certification body site, ai-certs.org, and the certification scheme still use the older level name. The next two levels were renamed the same way, to AI+ Security Expert and AI+ Security Strategist.
The AT-2101 exam has 50 single-response multiple-choice questions in a 90-minute slot, which AI CERTs splits into 80 minutes of exam time, 5 minutes for the candidate agreement and 5 minutes for the Proctor 365 tutorial. You need 70 percent, or 35 correct answers, to pass. Capstone and case study items show up as scenario-based multiple-choice questions, so there is no project to submit.
AI CERTs’ current AT-2101 blueprint has 12 modules. Computing, Linux and Operating System Foundations is 5 percent. Networking, Python, Cybersecurity Foundations, Cryptography and Identity, Introduction to AI and Machine Learning, AI for Detection and Threat Hunting, and AI and LLM Security are 9 percent each. Offensive Security for AI Systems, Security Operations and Incident Response, Governance and Compliance, and the Capstone are 8 percent each. The older scheme document on ai-certs.org still shows an 11-domain outline, so confirm with AI CERTs before you book.
No. The certification scheme sets no mandatory education, experience or prior-certification requirement, and training isn’t required either. AI CERTs recommends a basic understanding of operating systems and networking, introductory AI and cybersecurity concepts, and some programming awareness, with Python preferred but not required.
As of October 2026, AI CERTs’ Candidate Handbook lists the exam-only fee at $195 and the recertification exam at $95. The AI CERTs store sells the self-paced AI+ Security Practitioner package, course plus online proctored exam, for $495. Every voucher includes one free retake, and vouchers expire six months after purchase. Training Camp’s AT-2101 boot camp includes the exam voucher.
AI CERTs says it operates as an independent certification body under ISO/IEC 17024:2012 and designs its schemes in line with the principles of that standard. It does not claim ANAB accreditation. Its accreditation page says AI CERTs is “actively pursuing ANAB Accreditation.” If accreditation matters to your employer, check the current status with AI CERTs directly.
AI CERTs makes no DoD 8140 claim for AI+ Security Practitioner, so don’t count on it to meet a DoD 8140 position requirement. If you need a credential for a DoD cyber role, Security+ is approved for multiple DCWF work roles. See the current DoD 8140 work role paths.
They are aimed at different stages. AI+ Security Practitioner is foundational, has no prerequisites, uses 50 multiple-choice questions and renews every year by re-exam. SecAI+ (CY0-001) expects three to four years of IT experience, adds performance-based questions, is ANAB-accredited and lasts three years. Start with AI+ Security Practitioner if you’re new to security, or go to SecAI+ if you already work in it.
AI CERTs maps AT-2101 to roles including AI Security Analyst, Cybersecurity Analyst, SOC Analyst, Threat Hunter, Security Engineer, AI Risk Specialist and Incident Response Analyst. As a foundational credential, it works best as an entry into SOC and analyst work when paired with hands-on experience or a core certification like Security+.
AI CERTs certifications are valid for one year from issuance. The only renewal method is a secure, proctored multiple-choice re-exam taken before the certificate expires, and a pass renews it for another year. If you can’t make the deadline, you can ask for a grace period of 30 to 90 days before expiry. Miss both and the certificate is suspended as non-recertified.
It can be, for the right person. If you’re early in a security career and want AI covered alongside the basics, it’s a $195, online, 50-question exam with no prerequisites. The tradeoffs are a yearly re-exam, no ANAB accreditation claim and no DoD 8140 claim. If a DoD role is the goal, Security+ comes first. If you already work in security, SecAI+ is the deeper AI security exam.
Choosing between AI+ Security Practitioner and Security+, or planning AI security training for a whole team? Tell us where you are and we’ll help you map out the next step.