Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about the unique logging and observability challenges that arise when implementing AI-as-a-Service solutions in this 24-minute conference talk from fwd:cloudsec. Explore how Large Language Models (LLMs) and other Platform-as-a-Service offerings built from third-party services on cloud service providers create distinct monitoring difficulties compared to traditional infrastructure. Discover practical approaches and lessons learned for addressing these challenges from both observability and detection and response perspectives. Gain insights into the complexities of tracking and monitoring AI services that operate as black boxes, understanding the limitations of traditional logging methods when applied to AI workloads, and implementing effective strategies for maintaining visibility into AI-powered applications. Examine real-world scenarios where conventional monitoring falls short and learn proven techniques for establishing comprehensive logging frameworks that can handle the unique characteristics of AI services, including their unpredictable resource consumption patterns, complex data flows, and integration challenges with existing security and monitoring tools.
