Certified - AI Security Audio Course
The AI Security & Threats Audio Course is a comprehensive, audio-first learning series focused on the risks, defenses, and governance models that define secure artificial intelligence operations today. Designed for cybersecurity professionals, AI practitioners, and certification candidates, this course translates complex technical and policy concepts into clear, practical lessons. Each episode explores a critical aspect of AI security—from prompt injection and model theft to data poisoning, adversarial attacks, and secure machine learning operations (MLOps). It also covers global standards and regulatory guidance, including the NIST AI Risk Management Framework, ISO/IEC 23894, and emerging organizational policies around transparency, accountability, and continuous monitoring.
Episodes
Episode 1 — Course Overview & How to Use This Prepcast
This opening episode provides a structured orientation to the AI Security and Threats Audio course series, helping listeners understand what the program covers and how to best engage with the material. The overview defines the scope of AI security by placing it within the broader context of cybersecurity and risk management, while clarifying the distinctive elements that make AI-specific
Episode 2 — The AI Security Landscape
This episode defines the AI security landscape by mapping the assets, attack surfaces, and emerging threats that distinguish AI from classical application security. It introduces critical components such as training data, model weights, prompts, and external tools, explaining why each must be protected as an asset. The relevance for certification exams lies in understanding how these comp
Episode 3 — System Architecture & Trust Boundaries
This episode explains the architecture of AI systems, breaking down their stages and components to show how trust boundaries shift across the lifecycle. Training, inference, retrieval-augmented generation (RAG), and agent frameworks are introduced as discrete but interconnected environments, each with distinct risks. For exam relevance, learners are expected to identify these architectura
Episode 4 — Data Lifecycle Security
This episode examines data lifecycle security, covering the journey of data from collection and labeling through storage, retention, deletion, and provenance management. It explains why data is the foundation of AI system reliability and how its misuse or compromise undermines security objectives. For certification preparation, learners are introduced to key definitions of provenance, int
Episode 5 — Prompt Security I: Injection & Jailbreaks
This episode introduces prompt injection and jailbreaks as fundamental AI-specific security risks. It defines prompt injection as malicious manipulation of model inputs to alter behavior and describes jailbreaks as methods for bypassing built-in safeguards. For certification purposes, learners must understand these concepts as new categories of vulnerabilities unique to AI, distinct from
Episode 6 — Prompt Security II: Indirect & Cross-Domain Injections
This episode examines indirect and cross-domain prompt injections, which expand the attack surface by embedding malicious instructions in external sources such as documents, websites, or email content. Unlike direct injection, where the attacker provides inputs to the model directly, these threats exploit retrieval or integration features that feed information into the AI system automatic
Episode 7 — Content Safety vs. Security
This episode explains the distinction and overlap between content safety and security in AI systems, a concept often emphasized in both professional practice and certification exams. Content safety refers to filtering or moderating outputs to prevent harmful or offensive material, while security focuses on protecting systems and assets from adversarial manipulation or data loss. Although
Episode 8 — Data Poisoning Attacks
This episode introduces data poisoning as a high-priority threat in AI security, where adversaries deliberately insert malicious samples into training or fine-tuning datasets. For exam readiness, learners must understand how poisoning undermines model accuracy, introduces backdoors, or biases outputs toward attacker goals. The relevance of poisoning lies in its persistence, as compromised
Episode 9 — Training-Time Integrity
This episode covers training-time integrity, focusing on the assurance that data, processes, and infrastructure used in model development remain uncompromised. Learners preparing for exams must understand that threats at this stage include data tampering, corrupted labels, or manipulated hyperparameters. Unlike inference-time attacks, which target deployed models, training-time compromise
Episode 10 — Privacy Attacks
This episode introduces privacy attacks in AI systems, focusing on techniques that reveal sensitive or personal information from training data or model behavior. Learners must be able to define key attack types, such as membership inference—determining whether a specific record was included in training—and model inversion, where attackers reconstruct approximate training inputs. The exam
Episode 11 — Privacy-Preserving Techniques
This episode explores privacy-preserving techniques designed to reduce the risk of sensitive information exposure in AI systems while maintaining utility of the models. Learners must understand concepts such as anonymization, pseudonymization, and data minimization, which limit identifiable information in training sets. Differential privacy is introduced as a mathematical framework that i
Episode 12 — Model Theft & Extraction
This episode addresses model theft and extraction, highlighting how adversaries can replicate or steal valuable AI models. Model theft occurs when proprietary weights or architectures are exfiltrated, while model extraction involves querying an exposed API repeatedly to reconstruct decision boundaries or functionality. For exam purposes, learners must be able to distinguish between these
Episode 13 — Adversarial Evasion
This episode introduces adversarial evasion, a class of attacks in which maliciously crafted inputs cause AI systems to misclassify or behave incorrectly. For exam purposes, learners must be able to define adversarial examples, explain why they are often imperceptible to humans, and distinguish them from poisoning attacks, which occur during training. Evasion attacks take place at inferen
Episode 14 — RAG Security I: Retrieval & Index Hardening
This episode explores retrieval-augmented generation (RAG) security, focusing on retrieval and index hardening as foundational defenses. RAG combines language models with external document retrieval, which improves factual grounding but introduces risks. Learners preparing for exams must understand how poisoning of indexes, adversarial queries, and tampered retrieval sources can compromis
Episode 15 — RAG Security II: Context Filtering & Grounding
This episode continues exploration of RAG security by examining context filtering and grounding as defenses for reliable outputs. Learners must understand context filtering as the screening of retrieved documents before they are passed to a model, ensuring that malicious or irrelevant content is excluded. Grounding is defined as aligning model outputs to trusted sources, improving accurac
Episode 16 — Agents as an Attack Surface
This episode introduces AI agents as a new and growing attack surface, highlighting how their autonomy and tool integration create unique risks. Agents differ from single-response models by persisting through plan-and-act loops, chaining multiple steps, and invoking external tools or APIs. For certification purposes, learners must understand that these design features expand the system bo
Episode 17 — Secrets & Credential Hygiene
This episode addresses secrets and credential hygiene, emphasizing their critical role in preventing leaks and privilege misuse in AI systems. Secrets include API keys, tokens, passwords, and configuration values embedded in prompts or environments. Learners preparing for exams must understand that secrets frequently appear in AI workflows, often stored insecurely or accidentally revealed
Episode 18 — AuthN/Z for LLM Apps
This episode explores authentication (AuthN) and authorization (AuthZ) for large language model (LLM) applications, highlighting their importance in managing identities and permissions. Authentication verifies that a user or system is who they claim to be, while authorization defines what actions or resources they are allowed to access. For certification readiness, learners must understan
Episode 19 — Output Validation & Policy Enforcement
This episode examines output validation and policy enforcement as mechanisms for controlling what AI systems produce before results are delivered to users or downstream processes. Output validation ensures that responses conform to expected formats or structures, such as JSON schemas, while policy enforcement applies organizational rules that block disallowed or unsafe outputs. For exam p
Episode 20 — Red Teaming Strategy for GenAI
This episode introduces red teaming as a structured method for probing generative AI systems for vulnerabilities, emphasizing its importance for both exam preparation and real-world resilience. Red teaming involves adopting an adversarial mindset to simulate attacks such as prompt injection, data leakage, or abuse of system integrations. For learners, understanding red team goals, rules o
Episode 21 — Evals & Test Pipelines
This episode examines evaluations and test pipelines as essential processes for maintaining AI system security and reliability. Evaluations, or “evals,” are structured tests that measure a model’s behavior against known benchmarks or adversarial scenarios, while pipelines provide the automated flow of regression testing, scorecards, and service-level objectives. For certification purposes
Episode 22 — Telemetry & Observability
This episode explores telemetry and observability, emphasizing their importance in detecting anomalies, intrusions, and misuse in AI systems. Telemetry refers to the data collected from inputs, outputs, and system components, while observability describes the ability to interpret that data to understand system health and behavior. For exams, learners must know what to log, how to ensure l
Episode 23 — Abuse & Fraud Detection
This episode addresses abuse and fraud detection in AI applications, focusing on how adversaries exploit systems for spam, phishing, or marketplace manipulation. For certification purposes, learners must understand definitions of abuse, such as misuse of generative models for disallowed tasks, and fraud, defined as deceptive actions for financial or reputational gain. The exam relevance l
Episode 24 — Cost & Resource Abuse
This episode examines cost and resource abuse, where adversaries or careless users exploit AI systems to drive up compute expenses or deny service to legitimate customers. For exams, learners must understand the concept of denial-of-wallet, token-based denial-of-service, and quota storms that can overwhelm infrastructure. These risks are unique to AI because of resource-intensive inferenc
Episode 25 — MLOps & Serving Security
This episode introduces MLOps and serving security, focusing on practices that protect the deployment, operation, and continuous delivery of AI models. MLOps extends DevOps principles to AI, requiring controls for model registries, CI/CD pipelines, and serving infrastructure. For certification purposes, learners must know definitions such as model registry, rollback, and shadow deployment
Episode 26 — Supply Chain & Artifacts
This episode examines supply chain and artifact security, focusing on how external dependencies and stored components create systemic risks in AI systems. Artifacts include datasets, model weights, configuration files, and container images, each of which must be treated as high-value assets. For certification purposes, learners must be able to define supply chain risk in the AI context an
Episode 27 — Secure Fine-Tuning & Adaptation
This episode introduces secure fine-tuning and adaptation, explaining how customization of pre-trained models introduces both benefits and new risks. Fine-tuning adjusts model weights on domain-specific data, while adaptation methods such as LoRA, adapters, or reinforcement learning from human feedback (RLHF) provide lightweight ways to specialize behavior. For certification exams, learne
Episode 28 — API Gateways & Proxies for AI
This episode focuses on API gateways and proxies, emphasizing their role as critical control points for AI applications. An API gateway manages traffic to model endpoints, providing authentication, authorization, rate limiting, and policy enforcement. Proxies filter and shape requests or responses, enabling organizations to apply additional layers of validation. For certification purposes
Episode 29 — Code Execution & Sandboxing
This episode examines the risks of code execution in AI systems and the security benefits of sandboxing. Many AI applications incorporate features allowing generated or user-provided code to run, enabling advanced analysis, automation, or integration with development environments. For exam readiness, learners must define sandboxing as the isolation of execution environments to minimize ri
Episode 30 — Connector/Plugin Security
This episode addresses connector and plugin security, focusing on how third-party integrations expand the attack surface of AI applications. Connectors link systems to external data or services, while plugins extend model functionality by calling APIs or executing tasks. For certification, learners must be able to define these integration types and explain why they pose risks of privilege
Episode 31 — Cloud & Infra for AI
This episode explores cloud and infrastructure security in the context of AI, focusing on GPU clusters, multitenancy, storage, and network isolation. For certification readiness, learners must understand that AI workloads often demand specialized compute and storage, which in turn require hardened configurations and rigorous access controls. Misconfigurations in cloud services remain one
Episode 32 — Keys, Encryption & Attestation
This episode examines keys, encryption, and attestation as core mechanisms for ensuring confidentiality, integrity, and trust in AI systems. Keys form the foundation of cryptographic operations, and encryption protects data at rest and in transit, as well as sensitive model artifacts such as weights and parameters. Attestation provides proof that systems or hardware are running trusted co
Episode 33 — Governance & Acceptable Use
This episode introduces governance and acceptable use policies as organizational frameworks that guide secure and ethical AI adoption. Governance defines the processes, roles, and oversight structures for managing AI risks, while acceptable use policies establish clear boundaries on how AI systems may be applied. For certification purposes, learners must understand that governance integra
Episode 34 — Risk Frameworks in Practice
This episode examines risk frameworks for AI security, focusing on the NIST AI Risk Management Framework and ISO/IEC 42001. These frameworks provide structured approaches to identify, assess, mitigate, and monitor AI-specific risks across technical and organizational domains. For certification exams, learners must understand how these frameworks map to real-world controls and governance p
Episode 35 — Threat Modeling for AI
This episode covers threat modeling as a structured method for identifying and prioritizing risks in AI systems. Learners must understand the role of frameworks such as MITRE ATLAS, which catalog adversarial techniques, and STRIDE, which provides categories like spoofing, tampering, and information disclosure. For certification purposes, it is essential to define the steps of threat model
Episode 36 — OWASP GenAI/LLM Top 10
This episode introduces the OWASP GenAI/LLM Top 10, a structured list of the most critical risks associated with generative AI and large language models. For certification purposes, learners must understand how OWASP adapts its long-standing methodology for web applications to the AI context, focusing on vulnerabilities such as prompt injection, insecure output handling, training data poi
Episode 37 — Secure SDLC for AI
This episode examines the secure software development lifecycle (SDLC) for AI, emphasizing integration of security at each stage of system creation. Learners must understand that AI-specific risks require adapting traditional SDLC practices to include dataset vetting, model validation, and adversarial testing. For exams, candidates should know the differences between general secure develo
Episode 38 — Incident Response for AI Events
This episode addresses incident response for AI-specific security events, focusing on structured detection, containment, and remediation. Learners must understand that AI incidents differ from traditional security breaches because they involve unique assets such as models, prompts, and training datasets. Exam candidates should be familiar with phases of incident response adapted to AI, in
Episode 39 — Deepfakes & Synthetic Media Risk
This episode explores the risks of deepfakes and synthetic media, examining how generative AI enables the creation of realistic but deceptive audio, video, and images. For certification, learners must understand definitions of deepfakes, the technologies behind them such as generative adversarial networks and diffusion models, and the societal risks they introduce. Exam relevance includes
Episode 40 — Content Provenance & Watermarking
This episode examines content provenance and watermarking as methods to authenticate AI-generated or human-created content, providing assurance of originality and integrity. Provenance involves tracking the history and origin of digital assets, often through metadata or cryptographic proofs, while watermarking embeds identifiable signals into content to mark it as genuine. For certificati
Episode 41 — Legal & Compliance Horizon (High-Level)
This episode introduces the legal and compliance horizon for AI security, giving learners a high-level view of regulatory landscapes without overwhelming them with acronyms. For certification readiness, candidates must understand that laws and policies increasingly define how AI systems are designed, deployed, and monitored. The relevance lies in recognizing the broad trends: stricter dat
Episode 42 — Third-Party & Vendor Risk
This episode explores third-party and vendor risk management in AI security, focusing on the challenges of relying on external providers for models, datasets, APIs, and infrastructure. For certification purposes, learners must understand that external dependencies create systemic risks when suppliers fail to secure their assets or comply with regulations. Exam questions often emphasize su
Episode 43 — Enterprise Architecture Patterns
This episode examines enterprise architecture patterns for secure AI deployments, focusing on how organizations structure systems to balance scalability, performance, and resilience. For certification, learners must understand concepts such as zero-trust architecture, network segmentation, and tiered environments for development, testing, and production. The exam relevance lies in recogni
Episode 44 — People & Process
This episode focuses on people and process as integral elements of AI security, highlighting how organizational culture and defined responsibilities reinforce technical defenses. For certification purposes, learners must understand that even the best security tools fail without proper governance structures, training programs, and accountability models. The exam relevance lies in recognizi
Episode 45 — Program Management Patterns (30/60/90)
This episode introduces program management patterns for phased AI security adoption, with emphasis on the 30/60/90-day framework. For certification readiness, learners must understand how phased adoption reduces overwhelm, builds momentum, and ensures that AI security programs deliver measurable results. The exam relevance lies in demonstrating knowledge of structured approaches to govern
Episode 46 — Multimodal & Cross-Modal Security
This episode introduces multimodal and cross-modal security, focusing on AI systems that process images, audio, video, and text simultaneously. For certification readiness, learners must understand that multimodal systems expand attack surfaces because adversarial inputs may exploit one modality to affect another. Cross-modal injections—such as embedding malicious instructions in an image
Episode 47 — On-Device & Edge AI Security
This episode examines on-device and edge AI security, focusing on models deployed in mobile, IoT, or embedded systems where resources are constrained and connectivity may be intermittent. For certification purposes, learners must understand the unique risks of on-device AI, including theft of model files, tampering with local execution environments, and loss of centralized monitoring. The
Episode 48 — Guardrails Engineering
This episode covers guardrails engineering, emphasizing the design of policy-driven controls that prevent unsafe or unauthorized AI outputs. Guardrails include policy domain-specific languages (DSLs), prompt filters, allow/deny lists, and rejection tuning mechanisms. For certification purposes, learners must understand that guardrails do not replace security measures such as authenticatio
Episode 49 — Confidential Computing for AI
This episode introduces confidential computing as an advanced safeguard for AI workloads, focusing on hardware-based protections such as trusted execution environments (TEEs), secure enclaves, and encrypted inference. For exam readiness, learners must understand definitions of confidential computing, its role in ensuring confidentiality and integrity of model execution, and how hardware r
Episode 50 — Automated Adversarial Generation
This episode examines automated adversarial generation, where AI systems are used to create adversarial examples, fuzz prompts, and continuously probe defenses. For certification purposes, learners must define this concept and understand how automation accelerates the discovery of vulnerabilities. Unlike manual red teaming, automated adversarial generation enables self-play and continuous
Welcome to the AI Security Course
Recommended
The Joe Rogan Experience
World News Tonight with David Muir
Talk About Talk - Executive & Leadership Communication Skills
This Past Weekend w/ Theo Von
Stand In The Circle
Conspiracy Files with Paige Carter
Learn English B1 with Daily News | English Listening Practice
Bad Friends
The Swerve Podcast: Obscure Topics | Conspiracy Theories
The Bread and Banter Podcast
The Church of What's Happening Now: The New Testament
Deadline: White House
