Panel: Deploying Enterprise AI Agents Securely скачать в хорошем качестве

Panel: Deploying Enterprise AI Agents Securely

What does it really take to deploy AI agents securely at the enterprise level? In this insightful panel, four experts from leading institutions share lessons learned, risks observed, and strategies implemented when putting AI agents into real-world production environments. Panelists: Joyce Gordon – Head of AI at Amperity David Lowe – Responsible AI at Microsoft Tanu Mitra – Human-AI Alignment Professor at University of Washington Adam Loving – Partner Engineer for Llama Enterprise at Meta The discussion covers everything from prompt injection, data leakage, and cultural bias to real-time monitoring, threat modeling, and proactive auditing. Panelists share hard truths about the current maturity of agent security practices, challenges with monitoring multi-agent systems, and the need to treat agents more like dynamic software workers—with logging, access control, and behavioral guardrails. Enterprise AI Agents Summit 2025 in Seattle. Hosted by OpenPipe + AWS on June 13, 2025.