669 words
3 minutes
AI Red Teaming : Part - 1

This is my first blog in a series dedicated to learning about AI Red Teaming. I am focusing on getting as much hands on as I can with dedicated local models that I will be using for testing out my methodology and improving my pattern recognition skills.

[!INFO] Here are my system specifications in case anyone would like to follow along:

  • The system that I am utilizing is a Lenovo LOQ laptop with 16 GB of RAM and i5-13450HX processor with RTX-3050 (6 GB VRAM) which is running Windows 11. And I have tested this setup to be sufficient for running upto 8B parameter models with ollama without spilling into system RAM under load and causing noticeable latency.
  • The content material I am using for this is given in these github repositories: repository 1 repository 2
  • And along with this, I also be using the OWASP AI Red Teaming Guide. Link

AI Red Teaming: Foundations and Frameworks#

Transitioning into AI red teaming requires a fundamental shift in methodology. While traditional pentesting targets system access and disk persistence, AI red teaming targets model behavior, data integrity, and algorithmic decision-making.

This guide outlines the foundational concepts of machine learning and the primary attack vectors used to compromise them.

Part 1: Machine Learning Foundations#

The AI Hierarchy#

Artificial Intelligence terminology is strictly hierarchical:

  • Artificial Intelligence (AI): The overarching category of intelligent machines and systems.

  • Machine Learning (ML): A subset of AI where algorithms identify patterns and learn from data.

  • Deep Learning (DL): A specialized subset of ML utilizing multi-layered artificial neural networks.

Training Methodologies#

Machine learning models require specific training mechanisms to process data.

Learning TypeMechanismAnalogy
Supervised LearningLearns from labeled datasets.Showing labeled flashcards to a student to identify objects.
Unsupervised LearningIdentifies hidden structures in unlabeled data.Sorting a mixed box of blocks by shape or color without prior instructions.
Reinforcement LearningLearns via trial and error utilizing a reward/penalty system.Navigating a maze where dead ends trigger penalties and the exit triggers a reward.

Foundational Algorithms#

Understanding core mathematical models is required for effective vulnerability research.

AlgorithmPrimary FunctionConcept / Application
Linear RegressionPredicts continuous numerical values.Drawing a line of best fit through a scatterplot (e.g., forecasting network traffic volume).
Logistic RegressionPerforms binary classification by mapping data to probabilities (0 to 1).Separating spam from benign emails utilizing a defined decision boundary threshold.
Decision TreesExecutes logic via sequential binary (Yes/No) nodes.Firewall rules (e.g., If packet > 500kb → check IP → Block or Allow).
Support Vector Machines (SVM)Establishes an optimal hyperplane to separate distinct data classes.The “Kernel Trick”: Mapping complex 2D data into 3D space to slide a flat boundary between intertwined data points.

Part 2: AI Red Teaming Methodologies#

Threat Modeling Frameworks#

Structured frameworks are necessary for executing attacks and communicating risk.

  • MITRE ATLAS: The adversarial threat landscape for AI systems, functioning similarly to the MITRE ATT&CK framework.

  • Google SAIF: Secures AI ecosystems by mapping trust zones across the Model, Data, Application, and System infrastructure.

  • NVIDIA AI Kill Chain: Defines the operational lifecycle of an AI attack:

    ReconnaissancePoisoningHijackingPersistenceImpact

Primary Attack Vectors (OWASP ML Examples)#

The following vectors map traditional web vulnerabilities to their machine learning equivalents.

IDAttack VectorMechanismPractical Example
ML01Input Manipulation (Evasion)Modifying inputs at runtime to deceive the model’s classification.Placing black tape on a Stop sign. A human recognizes the sign; an autonomous vehicle’s classifier identifies a 70 km/h speed limit sign.
ML02Data PoisoningInjecting malicious data into the training set to establish a backdoor.Submitting benign CVs containing a hidden trigger word (“Synergy”) to an HR screening tool. The model learns to automatically approve future CVs containing the word.
ML03Model InversionReconstructing private training data directly from model outputs.Querying a medical diagnostic model repeatedly to reverse-engineer the patient Personally Identifiable Information (PII) it was trained on.
ML04Membership InferenceDetermining if specific data was utilized during model training.Exploiting model confidence metrics to mathematically verify that an executive’s private emails were included in a public chatbot’s training dataset.
ML05Model TheftExfiltrating proprietary model architecture, weights, or parameters.Programmatically querying a competitor’s paid API, logging the inputs/outputs, and utilizing the dataset to train a functional clone.
ML06AI Supply Chain AttacksCompromising upstream AI components, dependencies, or repositories.Uploading a pre-trained model to a repository containing a reverse shell hidden via Pickle deserialization within the model weights. Execution occurs immediately upon load.