Certified AI & Autonomous Systems Security Specialist (CAASS)

Certified AI & Autonomous Systems Security Specialist (CAASS) Certification Program by Tonex

 Certified AI & Autonomous Systems Security Specialist (CAASS) Certification Program by Tonex prepares professionals to defend intelligent systems across complex real world environments. Participants explore how AI models, autonomous decision engines, and generative AI pipelines introduce new attack surfaces from data collection through deployment. The program connects model behavior with system level risk so learners can speak confidently with engineers, security teams, and executives.

Practical frameworks show how to identify threat vectors, evaluate robustness, and align AI driven capabilities with safety and mission goals. Strong emphasis is placed on cybersecurity controls around training data, inference interfaces, and integrated autonomous platforms, helping organizations reduce exposure to misuse and compromise. By the end of the program, participants are equipped to evaluate AI architectures, challenge design assumptions, and recommend controls that improve resilience, trustworthiness, and accountability.

Learning Objectives

• Understand major threat categories across AI, ML, autonomy, and generative models throughout the system lifecycle
• Analyze training and inference pipelines to locate weaknesses in data handling, model exposure, and deployment workflows
• Evaluate model robustness using structured testing techniques, red teaming strategies, and adversarial scenarios
• Interpret bias, integrity, and trustworthiness metrics and translate findings into actionable risk treatment options
• Design model assurance documentation that supports auditability, compliance, and executive level decision making
• Connect AI security and trust engineering practices with broader cybersecurity governance, risk, and compliance efforts

Audience

• Cybersecurity Professionals
• AI and ML Engineers
• Autonomy and Robotics Architects
• Security Architects and Engineers
• Risk and Compliance Managers
• DevOps and MLOps Practitioners
• Technical Product and Program Managers

Program Modules

Module 1: Foundations of AI and Autonomy Security

• Taxonomy of AI and autonomous systems
• Threat landscape for intelligent systems
• Mapping AI components to attack surfaces
• Data pipelines and asset classification
• Trust, safety, and reliability concepts
• Roles and responsibilities across teams

Module 2: Model Threats, Training and Inference Risks

• Data poisoning and backdoor insertion paths
• Model theft, extraction, and cloning attacks
• Prompt injection and jailbreak style abuses
• Inference abuse, overreach, and query abuse
• Shadow models and uncontrolled reuse risks
• Monitoring signals for anomalous behavior

Module 3: Securing LLM, RAG and GenAI Systems

• RAG architectures and data exposure paths
• Prompt design, guardrails, and policy layers
• Content filtering and safety policy enforcement
• Hardening connectors, tools, and plug ins
• Protecting proprietary and personal data
• Observability for LLM and RAG pipelines

Module 4: Robustness, Evaluation and Assurance Evidence

• Robustness testing strategies and coverage
• Adversarial input generation and screening
• Stress testing for autonomy decision logic
• Reliability metrics and trust scoring models
• Model assurance documentation and claims
• Linking evaluations to business risk posture

Module 5: Bias, Integrity, Governance and Trust

• Sources of bias in data and labels
• Fairness, inclusion, and impact assessments
• Integrity controls for data and model assets
• Governance frameworks for AI decision systems
• Alignment with safety and ethical guidelines
• Reporting to boards, regulators, and partners

Module 6: AI Red and Blue Teaming Practices

• Building AI focused threat models and paths
• Red team planning, tooling, and execution
• Blue team playbooks and defensive patterns
• Incident response for AI and autonomy failures
• Continuous validation in engineering pipelines
• Metrics for effectiveness and improvement

Exam Domains

1. Foundations of AI and Autonomous Systems Security
2. Threat Modeling and Attack Techniques for Intelligent Systems
3. Secure AI Development, Deployment, and Data Pipelines
4. Evaluation, Red Teaming, and Continuous Monitoring
5. Governance, Ethics, and Regulatory Alignment in AI Security
6. Autonomous Systems Safety, Resilience, and Mission Assurance

Course Delivery
The course is delivered through expert led lectures, interactive discussions, structured case studies, and guided practical exercises that focus on real AI and autonomous system implementations. Participants work through scenarios that link model behavior to operational and business risk, while exchanging experiences with peers from security, engineering, and governance roles. Digital materials, reference frameworks, and reusable templates are provided to help participants apply the methods directly within their own organizations.

Assessment and Certification
Participants are assessed through quizzes, short written assignments, and an integrative capstone style evaluation focused on an AI or autonomous system scenario. Performance is measured on technical understanding, analytical depth, and the ability to translate findings into clear recommendations. Upon successful completion of the assessments, participants receive the Certified AI & Autonomous Systems Security Specialist (CAASS) Certification from Tonex.

Question Types

• Multiple Choice Questions (MCQs)
• Scenario-based Questions

Passing Criteria
To pass the Certified AI & Autonomous Systems Security Specialist (CAASS) Certification Training exam, candidates must achieve a score of 70% or higher.

Strengthen your organization’s defenses for AI, autonomy, and generative technologies by building deep, practical security expertise. Enroll your team in the Certified AI & Autonomous Systems Security Specialist (CAASS) Certification Program by Tonex and turn advanced AI capabilities into secure, trustworthy, and resilient assets.