Brian Vermeer - Breaching your LLM-Powered Java Applications

This talk explores the integration of large language models (LLMs) with databases and the role of intelligent agents in online purchasing, highlighting both the potential applications of AI and the associated security and data privacy risks. The speaker addresses critical issues such as prompt injection, key leakage, and the misuse of private customer data in model training, along with relevant legal restrictions. Attendees will gain a comprehensive understanding of the vulnerabilities to avoid and the strategies necessary to ensure data privacy compliance, as well as best practices for developing secure applications powered by LLMs.