Trustworthy AI in the Cloud: Observability as a Control System (Sept. 22nd)
Tuesday, September 22nd, 2026: 1:00 PM to 1:30 PM
This session explores why observability is emerging as a foundational capability for trustworthy AI. Rather than focusing solely on infrastructure health, modern observability provides the transparency and context needed to understand how AI systems behave, why they produce specific outcomes, and whether they continue to perform as intended over time.
Virtual
AI is rapidly becoming a core component of modern cloud architectures, powering decisions, automating workflows, and influencing business outcomes. As organizations scale AI adoption, they face a new challenge: establishing trust in systems that are dynamic, adaptive, and increasingly autonomous. Traditional approaches to governance and compliance were designed for deterministic software—not AI systems whose behavior can vary based on data, context, and user interactions. The ability to observe, evaluate, explain, and govern intelligent systems will become essential for building trust, meeting evolving compliance expectations, and enabling innovation with confidence.
Attendees will examine how observability enables traceability, explainability, and continuous AI governance across applications, data, infrastructure, and AI services. We'll discuss emerging practices such as LLM-as-a-Judge evaluation frameworks, runtime AI assessments, and automated feedback loops that help organizations measure quality, validate business outcomes, and improve confidence in AI-driven decisions. Cloud Security Alliance: CloudBytes
Hosted by CSA