Packt Adversarial AI Attacks, Mitigations, and Defense Strategies: A cybersecurity professional's guide to AI attacks, threat modeling, and securing AI with MLSecOps
Product ID: 737690409
Buy anything from 5,000+ international stores. One checkout price. No surprise fees. Join 2M+ shoppers on Desertcart.
Desertcart purchases this item on your behalf and handles shipping, customs, and support to Vietnam.
Buy Packt Adversarial AI Attacks, Mitigations, and Defense Strategies: A cybersecurity professional's guide to AI attacks, threat modeling, and securing AI with MLSecOps by Sotiropoulos, John online on desertcart.ae at best prices. ✓ Fast and free shipping ✓ free returns ✓ cash on delivery available on eligible purchase. Review: This is not just another rephrasing of security guidelines. The book walks you through the entire AI security spectrum - starting from first principles, assuming no prior knowledge, and progressively building real, hands-on examples. You build a dockerised application, a mobile app sample, and a bot, using them to explore the full range of AI threats with clear explanations and practical demonstrations. One standout section for me was visualising embeddings and detecting signs of poisoning - something rarely explained this well. What really sets this book apart is how it then transitions into secure-by-design AI thinking and threat modelling, grounded in NIST, MITRE, and OWASP taxonomies, and tied together with guardrails and a Jenkins DevSecOps pipeline. A genuinely practical, in-depth, end-to-end guide that bridges theory and real-world implementation. A true classic. Review: A necessary book for those who in work in the AI Security field. It is easy to read and easy to follow along with practice labs. The author did a good job incorporating the breadth of AI threats.
| Best Sellers Rank | #55,012 in Books ( See Top 100 in Books ) #94 in Networking & Cloud Computing #96 in Computer Security & Encryption #337 in Computer Science |
| Customer reviews | 4.8 4.8 out of 5 stars (34) |
| Dimensions | 19.05 x 3.45 x 23.5 cm |
| Edition | Standard Edition |
| ISBN-10 | 1835087981 |
| ISBN-13 | 978-1835087985 |
| Item weight | 1.02 Kilograms |
| Language | English |
| Print length | 602 pages |
| Publication date | 26 July 2024 |
| Publisher | Packt Publishing |
C**N
This is not just another rephrasing of security guidelines. The book walks you through the entire AI security spectrum - starting from first principles, assuming no prior knowledge, and progressively building real, hands-on examples. You build a dockerised application, a mobile app sample, and a bot, using them to explore the full range of AI threats with clear explanations and practical demonstrations. One standout section for me was visualising embeddings and detecting signs of poisoning - something rarely explained this well. What really sets this book apart is how it then transitions into secure-by-design AI thinking and threat modelling, grounded in NIST, MITRE, and OWASP taxonomies, and tied together with guardrails and a Jenkins DevSecOps pipeline. A genuinely practical, in-depth, end-to-end guide that bridges theory and real-world implementation. A true classic.
P**E
A necessary book for those who in work in the AI Security field. It is easy to read and easy to follow along with practice labs. The author did a good job incorporating the breadth of AI threats.
E**A
Muito bom!
S**A
A thorough and complete book. It manages to combine research papers, best practices, standards, and code samples in a long but an easy-to-follow journey. At the end, I felt confident to engage and work in this complex area.
T**Z
A very good book, I absolutely love how it gives security professionals much needed AI security insights.
Trustpilot
2 weeks ago
2 months ago