As AI agents become more autonomous, executives face a critical trust gap when deploying them for higher-stakes applications where incorrect outputs and actions carry significant consequences. Building trustworthy agents requires responsible AI best practices that guide how agents are built and operated, addressing considerations like privacy, fairness, transparency, and governance. Automated reasoning provides techniques to validate AI responses, check properties of generated code, or enforce policies for agentic behaviors, all using mathematical logic for provable accuracy. Join this interactive session to explore practical approaches for building more trustworthy and reliable AI agents.