Media Summary: AI systems don't always fail because they're hacked—sometimes they fail because they're trained on malicious Can AI be hacked into lying? Behind every powerful model is a hidden battlefield, where attackers craft prompts, AI agents are incredibly helpful—but that also makes them vulnerable. In this episode, we dive into prompt injection and
Data Poisoning Adversarial Attacks Explained - Detailed Analysis & Overview
AI systems don't always fail because they're hacked—sometimes they fail because they're trained on malicious Can AI be hacked into lying? Behind every powerful model is a hidden battlefield, where attackers craft prompts, AI agents are incredibly helpful—but that also makes them vulnerable. In this episode, we dive into prompt injection and AI is everywhere, but have you ever thought about how it can impact security? USENIX Security '22 - Poison Forensics: Traceback of Hackers don't always break into your systems with code — sometimes, they
The MLSecOps Podcast Season 1 Episode 2 With Guest Florian Tramér, PhD In this episode, we interview Florian Tramèr, PhD ...
