IBMcopy121: Securing AI on Z Addressing Emerging Threats and Building Trustworthy Systems

Project and Program: Security

Tags: Proceedings, SHARE Orlando 2026, 2026

AI is transforming how organizations make decisions and deliver value—but it also introduces new security challenges. Threats like data poisoning, adversarial manipulation, and model theft can compromise both business outcomes and customer trust. Without proper safeguards, these risks can undermine trust, compromise sensitive data, and expose organizations to compliance violations. In this session, we will examine the evolving security threats unique to AI and highlight why securing AI pipelines is as important as securing the data itself. We’ll explore best practices for protecting training data, safeguarding models in production, and embedding transparency and accountability into AI workflows.

Back to Proceedings File Library