Ottawa AI SecEng - October Edition - Prompt Injection Deep Dive

​Prompt Injection Deep Dive + AI Security CTF

​Prompt injection is one of the biggest security challenges for apps built on large language models (LLMs). AI systems now connect to documents, websites, internal data, APIs and external tools. That gives attackers more places to plant malicious instructions that change how the system behaves, and in some cases trigger actions nobody intended.

​Join AI Security Engineers Ottawa for a technical evening on how these attacks work, where they show up in real AI architectures, and how engineers can build more resilient AI applications.

​Agenda

• ​AI security news and practical engineering topics

• ​Prompt injection deep dive: how the attacks work, where they appear in real systems, and how to defend against them

• ​Beginner-friendly AI security capture the flag (CTF), with a $15 gift card if you solve a challenge (thanks, Snyk!)

• ​Pizza and soft drinks