From the course: Security Risks in AI and Machine Learning: Categorizing Attacks and Failure Modes
Unlock this course with a free trial
Join today to access over 24,900 courses taught by industry experts.
Supply chain attacks
From the course: Security Risks in AI and Machine Learning: Categorizing Attacks and Failure Modes
Supply chain attacks
- [Instructor] In the software development world, most applications aren't created entirely from scratch. When application developers need certain functionality, like showing a location on a map, they don't typically write an entire mapping system from the ground up. Instead, they embed one of the many popular mapping services or APIs into their application. In this way, developers speed up the delivery of their software by leveraging the work of others. This happens thanks to a robust and complex software supply chain. There's also a complex supply chain for AI development too. This takes the form of libraries and APIs that developers use to run AI models. But it can also come in more specialized form, like pretrained models, foundation models, and model zoos. Pretrained models are models available for download that come already trained on large datasets and are able to perform a specific function. A well-known example of…
Contents
-
-
-
-
(Locked)
Perturbation attacks and malicious input7m 26s
-
(Locked)
Poisoning attacks6m 54s
-
(Locked)
Reprogramming3m 30s
-
(Locked)
Physical domain: 3D adversarial objects3m 55s
-
(Locked)
Supply chain attacks4m 4s
-
(Locked)
Model inversion5m 19s
-
(Locked)
System manipulation4m 49s
-
(Locked)
Membership inference and model stealing4m 26s
-
(Locked)
Backdoors and existing exploits3m 45s
-
(Locked)
-
-
-