The Agentic Systems Framework

About the Event

Event Overview

The Agentic Systems Framework: Trust, Security, and Governance will examine the practical questions organisations face as AI gains greater agency: how much autonomy should AI have, what institutional foundations are required before deployment, how autonomous systems should be secured, and how accountability and meaningful human oversight can be preserved.

The practical challenge is not simply whether an AI agent can perform a task, but whether the surrounding organisation is prepared to deploy it responsibly: whether the data and technology foundations are ready, whether permissions and authority are clearly defined, whether the system can be monitored and contained, and whether meaningful human intervention remains possible.

Technology professional working with a laptop in a data centre

Agenda Overview

The event will follow a focused programme structure designed to move from strategic context-setting to three thematic discussions, followed by audience interaction and closing reflections.

01

Opening Remarks and Conference Context Setting

1 Moderator

02

Opening Keynote

The Agentic Systems Framework: Trust, Security, and Governance

03

Theme 1

AI Autonomy & Decision-Making

2 Panelists (15 minutes each)

04

Theme 2

Security & Resilience of Autonomous AI

2 Panelists (15 minutes each)

05

Theme 3

Governance, Accountability & Human Oversight

2 Panelists (15 minutes each)

06

Interactive Live Q&A / Audience Discussion

07

Closing Reflections and Key Takeaways

Speakers & Moderators

Conference Speakers & Panelists

Moderator

Conference Moderator

Nitin Dhavate

Data Privacy & AI Governance Head – Asia Pacific, Middle East & Africa (APMA) and Global Health

AI Governance
Data Privacy
Responsible AI
AI & Privacy Risk Management
Governance Programme Design & Implementation
Enterprise AI Governance

Nitin is a privacy, security and AI governance professional with a focus on planning, implementing and maintaining privacy and AI governance programs to enable business growth while managing privacy and AI risks.

Nitin is currently the Data Privacy and AI Governance Head for Novartis Asia Pacific, Middle East and Africa, across all the business functions and units. He also works on global initiatives at Novartis.

Theme 3

Governance, Accountability & Human Oversight

Cesar Said Rosales Torres

Founder & Director, OuiCheck

Proposed Topic

Governing Increasingly Autonomous AI: Rethinking the Infrastructure for Autonomy

How emerging forms of AI autonomy are challenging traditional approaches to governance and human oversight, and what new infrastructure for authority, attention, assurance, and accountability may be needed to scale them responsibly.

César Rosales is the Founder and Director of OuiCheck, an AI assurance and advisory firm specializing in regulatory compliance, risk management, responsible AI, and the evaluation of AI systems for organizations across Europe and Latin America. OuiCheck helps organizations translate emerging AI regulation and responsible AI frameworks into practical approaches for managing risk, demonstrating trustworthiness, and scaling AI responsibly.

Over the past six years, he has served as an external advisor to the Inter-American Development Bank’s innovation laboratory, IDB Lab, on responsible AI, innovation, and venture ecosystems, where he coordinated and co-led fAIr LAC, an initiative focused on advancing the responsible adoption of AI among entrepreneurs, startups, and venture capital funds across Latin America and the Caribbean. He currently serves as an external consultant to the Inter-American Development Bank’s Technology and Transformation Department (TTD), focusing on AI governance and the responsible adoption of AI.

His work and research have also focused on algorithmic discrimination, AI governance, and the impact of emerging technologies on fundamental rights. He currently teaches on the impact of technology on fundamental rights in the Master in Technology and Global Affairs at IE University in Spain.

César holds a Master’s degree in Latin American Studies from Georgetown University, with a concentration in Political Economy, and a Bachelor’s degree in International Relations from Tecnológico de Monterrey in Mexico.

John M. Willis

Founder & Chief Innovation Officer, Sustainable Future Tech, Inc.

Proposed Topic

Beyond Guardrails: Governing Agentic AI at the Point of Action

The presentation will examine why governing agentic AI requires more than behavioral guardrails and policy declarations. It will introduce deterministic runtime governance as the means of establishing enforceable authorization, controlling consequential actions, incorporating human oversight where required, and preserving independently verifiable evidence of governance decisions and execution.

John M. Willis is Founder and Chief Innovation Officer of Sustainable Future Tech, Inc., and a cybersecurity architect, AI governance researcher, and author specializing in the secure engineering of autonomous systems.

He is the creator of Runtime Governance Engineering, an approach to translating governance requirements into enforceable controls over consequential actions performed by AI agents and other automated systems.

His research focuses on deterministic authorization and enforcement at the point of execution, including agent identity, authority, policy constraints, system invariants, evidence integrity, and accountable human oversight. He is the principal architect of the AI Governance Control Plane (AGCP), a framework for separating probabilistic AI reasoning from authoritative execution decisions.

His publications include Runtime Governance Engineering: Foundations and Architecture—From Governance Intent to Controlled Execution, along with research on runtime execution governance, formal execution semantics, and secure AI architecture.

His professional background spans enterprise cybersecurity architecture, federal cybersecurity programs, systems engineering, and emerging technologies. His work emphasizes moving AI governance beyond documented policies and behavioral guardrails toward independently verifiable, enforceable controls over consequential system behavior.

Expected Outcomes

The event is expected to:

The Agentic Systems Framework will create a practical bridge between AI capability and responsible institutional implementation. By bringing together leaders from technology, cybersecurity, governance, policy, research, and the public sector, the event will help clarify what organisations need in order to deploy increasingly autonomous AI systems with appropriate trust, security, accountability, and human oversight.

build a clearer shared understanding of how Agentic AI differs from conventional assistive AI and what greater autonomy means in operational settings;
identify practical readiness gaps organisations may encounter when moving Agentic AI from experimentation into operational environments;
improve understanding of security approaches for agent identity, permissions, runtime monitoring, containment, recovery, and operational resilience;
surface emerging governance approaches for accountability, decision authority, human approval, auditability, and continuous assurance;
highlight implementation experiences and practical lessons that can inform responsible Agentic AI adoption across sectors;
strengthen dialogue between technology, security, governance, policy, research, and public-sector stakeholders on the responsible deployment of autonomous AI systems;
support continued knowledge exchange and potential follow-up collaboration around trustworthy, secure, and governable Agentic AI.

Event Overview

Why Agentic AI Matters

Conference Objectives

From assistance to action

Autonomy and decision authority

Institutional foundations

Security and resilience

Accountability and intervention

Lessons from real deployments

Cross-sector dialogue

Agenda Overview