Talks·52 min
Rico Komenda · Javan Rasokat
Pairs a builder and a penetration tester to show LLM prompt injection attacks alongside layered defenses for AI-integrated apps.
aiappsecvulnerabilities+2
Rico Komenda
2 talks in the archive.
Filters+
1–2 of 2 talks
Talks·42 min
Rico Komenda
Maps the Model Context Protocol's attack surface, including tool poisoning, rug pulls and authorization flaws, with defensive measures.
aivulnerabilitiesattack+2