PracticeVideosBlogFlashcardsEspañol
All Practice Exams

100+ Free CAICSO Practice Questions

Pass your Mile2 Certified AI Cybersecurity Officer (CAICSO) exam on the first try — instant access, no signup required.

✓ No registration✓ No credit card✓ No hidden fees✓ Start practicing immediately
Not published Pass Rate
100+ Questions
100% Free
1 / 100
Question 1
Score: 0/0

Under the NIST AI RMF MANAGE function, which activity addresses the situation where an identified AI risk cannot be fully mitigated but is acknowledged and documented with a risk acceptance decision?

A
B
C
D
to track
Same family resources

Explore More Mile2 Cybersecurity Certifications

Continue into nearby exams from the same family. Each card keeps practice questions, study guides, flashcards, videos, and articles in one place.

Mile2 Certified Cloud Security Officer (CCSO)

Practice Questions

Mile2 Certified Digital Forensics Examiner (CDFE)

Practice Questions

Mile2 Certified Disaster Recovery Engineer (CDRE)

Practice Questions

Mile2 Certified Information Systems Risk Manager (CISRM)

Practice Questions

Mile2 Certified Information Systems Security Officer (CISSO)

Practice Questions

Mile2 Certified Network Forensics Examiner (CNFE)

Practice Questions
2026 Statistics

Key Facts: CAICSO Exam

100 MCQ

Exam Length

Mile2

70%

Passing Score

Mile2

20 modules

Course Modules

Mile2

Sep 2025

Course Release

Mile2

~2 hours

Exam Duration

Mile2

NIST AI RMF + EU AI Act

Core Frameworks

Mile2 / NICCS

The Mile2 C)AICSO is a 100-question online exam requiring 70% to pass, covering AI cybersecurity officer competencies across risk frameworks (NIST AI RMF, ISO 42001), regulatory compliance (EU AI Act), AI threat modeling (MITRE ATLAS, OWASP LLM Top 10), ML pipeline security, and AI incident response. Released September 2025.

Sample CAICSO Practice Questions

Try these sample questions to test your CAICSO exam readiness. Each question includes a detailed explanation. Start the interactive quiz above for the full 100+ question experience with AI tutoring.

1Which four core functions make up the NIST AI Risk Management Framework (AI RMF 1.0)?
A.Govern, Map, Measure, Manage
B.Identify, Protect, Detect, Respond
C.Plan, Do, Check, Act
D.Assess, Authorize, Monitor, Retire
Explanation: The NIST AI RMF 1.0, published January 2023, is organized around four core functions: Govern, Map, Measure, and Manage. Govern establishes organizational culture and processes; Map contextualizes AI risks; Measure analyzes and benchmarks risks; and Manage addresses risk response and treatment.
2Under the EU AI Act, which risk category covers AI systems that pose a threat to fundamental rights or safety and are therefore BANNED outright?
A.High-risk
B.Limited-risk
C.Unacceptable-risk
D.Minimal-risk
Explanation: The EU AI Act establishes four risk tiers. The 'unacceptable risk' category contains AI practices that are prohibited entirely because they pose a clear threat to safety, livelihoods, or fundamental rights. Examples banned from February 2025 include social scoring by public authorities, real-time remote biometric identification in public spaces for law enforcement, and systems that exploit psychological weaknesses.
3An attacker supplies a crafted user query that overrides an LLM's system prompt and causes it to exfiltrate internal data. Which AI-specific attack technique does this exemplify?
A.Prompt injection
B.Model inversion
C.Data poisoning
D.Model extraction
Explanation: Prompt injection exploits LLMs by embedding malicious instructions in inputs that override system-level directives, potentially causing the model to leak sensitive data, bypass safety filters, or perform unauthorized actions. It is catalogued in MITRE ATLAS and is the top risk in the OWASP Top 10 for LLM Applications.
4Which MITRE framework specifically catalogs adversarial techniques targeting machine learning and AI systems, including tactics such as ML supply chain attacks and model evasion?
A.MITRE ATLAS
B.MITRE ATT&CK Enterprise
C.MITRE D3FEND
D.MITRE CAPEC
Explanation: MITRE ATLAS (Adversarial Threat Landscape for AI Systems) is the dedicated knowledge base for adversarial ML tactics and techniques. As of version 5.1.0 it contains 16 tactics, 84 techniques, and 56 sub-techniques specifically mapped to AI attack surfaces such as training pipelines, inference APIs, model files, and vector stores.
5An organization wants to ensure their ML pipeline enforces access controls on the model registry and cryptographically signs model artifacts before deployment. Which security objective does this primarily address?
A.AI supply chain integrity
B.Model availability
C.Inference-time prompt safety
D.Regulatory audit logging
Explanation: Controlling access to the model registry and signing model artifacts addresses supply chain integrity — ensuring that only authorized, authentic model versions are deployed and that backdoored or tampered models cannot enter production. This is a key DevSecMLOps control aligned with NIST AI RMF and EU AI Act technical documentation requirements.
6Under the NIST AI RMF, the GOVERN function differs from the other three functions because it:
A.Applies across all phases and establishes organizational culture and accountability
B.Applies only to the design phase of the AI lifecycle
C.Replaces the Map function for high-risk AI systems
D.Focuses exclusively on regulatory compliance documentation
Explanation: GOVERN is distinctive in that it applies to ALL phases of an organization's AI risk management lifecycle, not just one stage. It cultivates organizational risk culture, assigns roles and responsibilities, establishes policies, and ensures leadership commitment — creating the foundation on which Map, Measure, and Manage operate.
7A malicious actor inserts fabricated medical records into a hospital's AI training dataset to cause the model to misclassify certain patient diagnoses in the attacker's favor. This is best described as:
A.Data poisoning
B.Evasion attack
C.Model inversion
D.Adversarial reprogramming
Explanation: Data poisoning involves injecting malicious or biased data into the training pipeline to corrupt a model's decision-making at inference time. The attacker does not need access to the model itself — only to the training data — and the corrupted model will behave as intended by the attacker once deployed.
8ISO/IEC 42001:2023 establishes requirements for which type of management system?
A.Artificial Intelligence Management System (AIMS)
B.Information Security Management System (ISMS)
C.Privacy Information Management System (PIMS)
D.Quality Management System (QMS)
Explanation: ISO/IEC 42001:2023 is the first international standard dedicated to an Artificial Intelligence Management System (AIMS). It provides a framework for establishing, implementing, maintaining, and improving responsible AI use, covering governance, risk, transparency, accountability, and bias mitigation across the AI lifecycle.
9During a RAG (Retrieval-Augmented Generation) deployment review, the security team discovers that users can retrieve documents belonging to other tenants via crafted queries. Which vulnerability class does this represent?
A.Prompt injection
B.Insufficient logging and monitoring
C.Vector and embedding weakness with tenant isolation failure
D.Excessive agency
Explanation: OWASP's Top 10 for LLM Applications identifies 'vector and embedding weaknesses' as a distinct risk, which includes failures to enforce tenant isolation in RAG systems. When access controls are not applied at the embedding or retrieval layer, adversarial similarity searches can retrieve cross-tenant documents, leaking sensitive information.
10Which EU AI Act provision requires providers of high-risk AI systems to register their systems in an EU-wide database before placing them on the market?
A.Article 9 — Risk management system
B.Article 13 — Transparency and provision of information
C.Article 51 — Registration of high-risk AI systems
D.Article 62 — Serious incident reporting
Explanation: Article 51 of the EU AI Act mandates that providers of high-risk AI systems covered by Annex III register their systems in a publicly accessible EU database maintained by the Commission before placing the system on the market or putting it into service. This registration supports transparency and regulatory oversight.

About the CAICSO Exam

The Mile2 CAICSO certifies professionals in the governance, risk management, and security of AI and ML systems. Released in September 2025, it covers 20 modules spanning AI fundamentals, NIST AI RMF, EU AI Act compliance, MITRE ATLAS-based threat modeling, securing ML pipelines, AI governance program design, auditing, and AI-centric incident response.

Questions

100 scored questions

Time Limit

Approximately 2 hours

Passing Score

70% (70/100)

Exam Fee

Varies by package (Mile2 Cybersecurity Institute)

CAICSO Exam Content Outline

~10%

AI Fundamentals and Architecture

AI/ML/GenAI concepts, AI system architecture, and business applications across sectors

~15%

Ethical, Legal, and Regulatory Compliance

EU AI Act risk categories, prohibited practices, high-risk obligations, ISO 42001, GDPR privacy-by-design

~15%

AI Risk Management Frameworks

NIST AI RMF (GOVERN/MAP/MEASURE/MANAGE), Mile2 Progressive AI RMF, risk assessment and appetite

~15%

AI Threat Modeling and Attack Landscape

MITRE ATLAS tactics, OWASP LLM Top 10, prompt injection, data poisoning, model extraction, neural backdoors

~15%

Securing AI Systems and ML Pipelines

Guardrails, RAG security, model signing, differential privacy, MLSecOps controls, cloud-native AI security

~10%

AI Governance and Organizational Program

AI governance frameworks, ethics committees, acceptable use policies, AI asset inventories, risk appetite

~10%

Auditing and Testing AI Systems

AI red teaming, bias evaluations, fairness metrics, model lifecycle governance gates, continuous monitoring

~10%

AI-Centric Incident Response

AI incident types, IR lifecycle, kill switch containment, rollback, canary recovery, post-incident review

How to Pass the CAICSO Exam

What You Need to Know

  • Passing score: 70% (70/100)
  • Exam length: 100 questions
  • Time limit: Approximately 2 hours
  • Exam fee: Varies by package

Keys to Passing

  • Complete 500+ practice questions
  • Score 80%+ consistently before scheduling
  • Focus on highest-weighted sections
  • Use our AI tutor for tough concepts

CAICSO Study Tips from Top Performers

1Know the four NIST AI RMF core functions: GOVERN (culture/accountability), MAP (context/risk identification), MEASURE (analysis/benchmarking), MANAGE (risk treatment/monitoring)
2Memorize EU AI Act risk tiers: Unacceptable (banned) → High-risk (Annex III obligations) → Limited-risk (transparency) → Minimal-risk (unregulated)
3MITRE ATLAS has 16 tactics and 84 techniques as of v5.1.0 — know Reconnaissance, ML Supply Chain Compromise, and Impact tactics with key AI-specific techniques
4OWASP LLM Top 10: LLM01 Prompt Injection, LLM06 Excessive Agency, LLM07 System Prompt Leakage, LLM08 Vector/Embedding Weaknesses are most exam-relevant
5Distinguish AI safety (unintended harms from design flaws) from AI security (deliberate adversarial attacks) — both are CAICSO exam topics
6EU AI Act application timeline: Feb 2025 (prohibitions + AI literacy), Aug 2025 (GPAI), Aug 2026 (full high-risk obligations)

Frequently Asked Questions

What is the Mile2 CAICSO exam format?

The Mile2 CAICSO exam consists of 100 multiple-choice questions with a 70% passing score (70 correct answers required). The exam takes approximately 2 hours and is delivered online through Mile2's Learning Management System. Standard and ANSI/DoD 8140 proctored formats are available.

What domains does the Mile2 CAICSO cover?

The CAICSO covers 20 modules organized around AI fundamentals, ethical and regulatory compliance (EU AI Act, ISO 42001), AI risk management frameworks (NIST AI RMF), AI threat modeling (MITRE ATLAS, OWASP LLM Top 10), securing AI and ML pipelines, AI governance program design, auditing and testing AI systems, AI-centric incident response, and AI resilience and futureproofing.

Who should pursue the Mile2 CAICSO?

The CAICSO targets CISOs, AI security program leads, risk managers, compliance officers, security architects, and senior security professionals responsible for governing, securing, and auditing AI systems within their organizations. It is designed for practitioners who must bridge technical AI security and organizational governance.

How does the CAICSO relate to the NIST AI RMF and EU AI Act?

The CAICSO body of knowledge is built around applying established AI governance and risk frameworks in practice. The NIST AI RMF (GOVERN, MAP, MEASURE, MANAGE) and EU AI Act (risk tiers, prohibited practices, high-risk obligations, FRIA, incident reporting) are core examination topics. Candidates must understand how to apply these frameworks to real AI deployments.

What AI-specific threats does the CAICSO exam cover?

The exam covers prompt injection, indirect prompt injection, data poisoning, model extraction (model stealing), neural backdoor attacks, adversarial examples (evasion attacks), membership inference, model inversion, jailbreaking, and AI supply chain attacks (foundation model compromise, ML library poisoning). MITRE ATLAS and OWASP LLM Top 10 are referenced frameworks.

How do I prepare effectively for the Mile2 CAICSO?

Focus on four pillars: (1) NIST AI RMF — understand GOVERN/MAP/MEASURE/MANAGE functions and what activities each contains; (2) EU AI Act — master the four risk tiers, prohibited practices list, high-risk obligations (Annex III, Annex IV technical documentation, Article 62 reporting); (3) AI threats — memorize MITRE ATLAS and OWASP LLM Top 10 categories with examples; (4) AI security controls — understand guardrails, differential privacy, model signing, RAG security, and MLSecOps pipeline controls.