Amidst severe manpower shortages and rising case complexities, Hong Kong's NGO sector faces unprecedented systemic pressure. This opening presentation introduces Agentic AI as a transformative paradigm shift—moving from passive software to autonomous, reasoning digital co-pilots that align with strong government encouragement and funding initiatives to modernize social services. The session contrasts the massive operational opportunities this technology brings to NGOs, such as automated administrative relief and scaled outreach, against its inevitable challenges, specifically data privacy risks, ethical governance, and the preservation of human empathy. Ultimately, this framework underscores that navigating the exact balance between AI opportunity and risk is the core mandate of today's event.

This session features a pioneering technology partner who is actively collaborating with local non-profits to co-create and deploy Agentic AI solutions, also covering the challenges of switching from experimentation to production ready. It explores real-world use cases, uncovers the practical challenges of frontline adoption, and provides actionable advice for NGOs looking to implement autonomous technology sustainably.

This session introduces the unique cybersecurity paradigm shift required for Agentic AI. Unlike traditional digital services built on predictable code, autonomous agents possess a non-deterministic, self-thinking nature that presents entirely new operational risks. This high-level briefing provides an overview of the vital protection frameworks NGOs must consider before deployment. Key topics introduce the core concepts of enforcing central governance over agent autonomy, implementing strict identity governance, adopting Zero Trust access control, and establishing central visibility to safely monitor unpredictable AI Agent actions in real time.

This session demystifies the critical cybersecurity risks associated with autonomous AI agents by breaking down complex technical vulnerabilities into clear, layman’s terms for NGO executives. Attendees will explore the core concepts of "Zero Trust” and learn how traditional security measures fail to protect against self-thinking technology. The presentation provides a practical, non-technical blueprint on how to apply identity-aware access controls to individual AI agents, ensuring that every autonomous action is strictly verified and monitored to safeguard sensitive community data and organizational assets.

This closing session brings NGO executives back to a sobering reality: how easy it is to attack non-profits in the era of AI. While Zero Trust Architecture (ZTA) successfully safeguards autonomous agents deployed in-house, it is not a silver bullet on its own. Attendees will examine why NGOs must ensure their existing digital services are being protected against sophisticated external threats. This briefing delivers a holistic security reality check, demonstrating how to build robust digital services protection to build a truly resilient organization.