Este programa pretende aportar un conocimiento de punta de lanza en el uso de Inteligencia Artificial en Ciberseguridad, para lo cual las personas asistentes aprenderán a utilizar las capacidades del Machine Learning, el Deep Learning, los Cognitive Services o la IA Generativa, adaptadas para mejorar sus recursos de Ciberseguridad.
Durante 45 horas, los estudiantes aprenderán las tecnologías más avanzadas de IA para mejorar sus habilidades hacia un nuevo mundo de Tecnologías IA-infused, en este caso, para convertirse en un profesional de la Ciberseguridad enriquecido con habilidades y destrezas de IA.
Adquirirán conocimientos avanzados en Machine Learning, Deep Learning, y Servicios Cognitivos específicos para aplicar en la mejora de las capacidades de ciberseguridad, incluyendo la detección de amenazas y la prevención de ataques.
Aprenderán a analizar, interpretar y codificar scripts para la creación de algoritmos de IA Generativa (GenAI), hardening y hacking, fundamentales para proyectos ofensivos y defensivos en ciberseguridad.
Desarrollarán proyectos utilizando modelos de IA como Keras para resolver problemas de ciberseguridad, incluyendo la detección de phishing, anomalías en redes, y comportamientos inusuales en sistemas operativos.
Aprenderán técnicas para el uso ofensivo y defensivo de la IA, incluyendo la creación y detección de DeepFakes, el manejo de modelos de lenguaje para pentesting y la protección contra ataques adversarios en aprendizaje automático.
Campus Bilbao
Otros
Inteligencia Artificial para Expertos en Ciberseguridad
Español
Online
Proceso de ingreso abierto
La Inteligencia Artificial ha cambiado drásticamente la Ciberseguridad, tanto para los atacantes, como para los defensores.
La IA Generativa ha hecho aparecer un nuevo frente de batalla de Ciberseguridad para el que no estamos preparados.
0
This module is intended to setup the fundations to run tools, frameworks and AI models coding in Python the algorithms that are going to be created in this course. Students will have the knowledge to analyse, interpret, transform and code scripts and hacking, hardening and developing GenAI algortithms that are going to be key in the rest of the classes. The practical exercises encourage an in-depth understanding of how to protect against and simulate cyber-attacks using AI technologies, preparing students for advanced challenges in cyber defense.
1
This foundational module is crucial for laying a solid groundwork in cybersecurity, focusing on privacy, security, and safety. It explores offensive and defensive security principles, crucial for navigating cybersecurity's complexities. By examining vulnerabilities, exploits, frameworks, and cryptography, students learn to identify and mitigate threats, essential for safeguarding information and system integrity. This module also introduces the integration of AI in cybersecurity, including Machine Learning, Deep Learning, and Generative AI; technologies that are not only tools for enhancing security strategies but also represent emerging fields that offer new solutions to cybersecurity challenges.
2
This module will provide the basics of Artificial Intelligence, starting with expert systems, and moving towards Machine Learning, Deep Learning, Cognitive Services, GANS and GenAI models. Students will carry out projects with Keras, to gain confidence in executing AI algorithms to solve different types of problems using ML, DL and Reinforcement Learning models. This approach not only consolidate their understanding of AI's theoretical aspects but also boosts their practical skills in employing AI algorithms across various scenarios, including data analysis and predictive modeling.
3
In this module, Cybersecurity and AI will be merged in several projects. Students will learn how to use AI to do Threat Hunting, to detect Spam Phishing and Spear Phishing attacks, using Machine Learning and Deep Learning algorithms.
At the end of this module, students will use ML and DL models to detect network anomalies and abnormal behaviour in operating systems to detect attackers in the enterprise using AI. In addition, students will review and analyse the capabilities of AI in Cybersecurity such as Eye Tracking, Facial Recognition, Biometric Identity, Machine Vision, etc., to improve security systems with Cognitive Services.
4
This module will cover all topics on Digital Humans or Synthetic Humans. In this module students will learn about DeepFakes, understanding how Cognitive Services work, how to use a GAN architecture to do lip sync, or face swapping, and how to use them in an APT against individuals or companies. At the same time, students will learn how Deep Fake detection techniques work to discover an adversary using a DeepFake stream in real time, or a DeepFake video to evade a KYC process or to generate a Fake News attack, using AI models. Students will use AI models to analyse heartbeat, blink, head position, eye tracking, eye glare, voice-based biometrics, etc., to detect DeepFakes.
5
This module will focus on Language Models, to explain how Large Language Models and Small Language Models can be used to generate DigitalIA-Infused Services, and all the issues related to them.
Students will use LLMs like GPT4, Llama or Gemini, on platforms like Bard or ChatGPT to understand how to look for vulnerabilities, how to use them in pentesting, how they can be used to infuse AI into Digital Humans, how they can affect identity, data poisoning, data breaches or privacy, etc. This is a module to use LLM/SLM as a powerful tool in Cybersecurity teams.
6
This module will cover the OWASP Top 10 for LLM Applications and Services, to analyse the robustness of a service or an application that is using LLMs. Students will learn about Prompt Injection, Hallucinations, Data Leaks, Client-Side Attacks using LLMs, how to attack a company using insecure plugins or jailbreaking techniques.
This module will analyse the RAG architecture for GenAI Digital Services to understand the concept of Augmented Prompting, Grounding or how to hack and protect LLMs using LLMs. It will also cover topics related to LLM firewalling and content security.
In the end, students will learn about the risks in self-created LLMs in enterprises, through Fine-Tuning, and the risks related to them.
7
This module will explain how Machine Learning based attacks work, such as Deep Exploiting, or FSGM and FGSM. Students will learn how to use the Open Machine Learning Security Project (OMLASP) to analyze the security of ML systems.
8
This module explains tools, procedures and frameworks for auditing, pentesting and monitoring AI-infused apps and services. Students will work with frameworks to assess the robustness, effectiveness and quality of AI models. In it, students will also study what are the guidelines for doing this work.
Todos los estudiantes tendrán como parte de su entrenamiento:
Cualquier persona interesada en este campo perteneciente al ámbito de la informática y con conocimientos básicos de seguridad, bien con titulación universitaria del campo de la informática, bien con titulación de FP de informática, o bien con experiencia de al menos tres años en puestos de administración de sistemas.
Para recibir más información puedes contactar con:
MENÚ 
