#NullconBerlin2025 | LLMs everywhere: The future is now and the past keeps repeating itself by Joern

In this talk, Joern Schneeweisz explores the pervasive integration of large language models (LLMs) into software applications and the implications this trend poses for security. He presents alarming real-world scenarios involving prompts, injections, and AI agents that intersect with traditional application security vulnerabilities. Additionally, Schneeweisz discusses safeguards and mitigation strategies to prevent catastrophic outcomes, such as compromised production databases, stemming from careless implementation of LLMs.