Can You Trust Your AI?

Explore the critical topic of trustworthy AI in this 46-minute conference talk from Devoxx. Delve into the world of Explainable AI (XAI) and its importance in making AI systems more transparent and reliable, especially in high-impact domains like healthcare and finance. Learn about the TrustyAI initiative at Red Hat, which focuses on enhancing decision trustworthiness through explainability, runtime tracing, and accountability. Discover various types of explanations, including local and knowledgeable explanations, and understand the importance of minimizing complexity while adhering to scientific methods. Examine concepts such as fairness, monitoring, and microservices in AI systems. Gain insights into practical applications, theoretical work on explainability, and the extraction of explanations from AI models. Conclude with a Q&A session addressing the significance of explainability in modern AI development and deployment.