Trusted Data for AI and Agentic Workflows

In this edition of STACK Meetup [Data], explore how you can safely unlock the value of sensitive data in an era of rapidly growing AI adoption and agentic workflows.

As organisations rapidly adopt AI and share data across increasingly complex ecosystems, new risks emerge, from data leakage and context exposure to unintended use of sensitive information. Join GovTech, IMDA, and Terminal 3 at this STACK Meetup as we explore how Responsible Data Use, Privacy-Enhancing Technologies (PETs), and trusted AI infrastructure enable secure AI adoption - from model development to cross-organisation collaboration and agentic AI systems.

Discover how GovTech enables Responsible Data Use by assessing data risk and selecting appropriate protections across the AI lifecycle. Learn how techniques such as anonymisation, synthetic data generation, and federated learning can support model development, cross-institution collaboration, and AI adoption at scale.

Take the opportunity to learn how IMDA's PET Sandbox provides practical tools supporting secure data collection, analysis, and sharing by protecting confidentiality and trust with strong governance and disciplined implementation. Get your hands on insightful adoption resources and technical guides while hearing about how to participate in future sandbox projects.

Deep dive with Terminal 3 on why safeguarding AI agentic systems requires hard-gating, not just software guard-railing. Understand how Trusted Execution Environments (TEEs) and Zero-Knowledge Proofs (ZKPs) can help protect sensitive data and enforce trust at use by blocking AI agents from simply reasoning around safety measures coded into software.

Whether you're an AI engineer, data practitioner, governance and policymaker, technology leader or data owner, you'll gain practical insights into selecting the right privacy-enhancing techniques, implementing responsible AI, and building trusted AI systems that can safely operate at scale for your organisation.