Uploaded November 2026 | Updated November 2026, 3 hours ago
Securing sensitive enterprise data as AI agents access information and execute business tasks requires consistent permissions, strong database protection, and manageable security policies. Explore how Sphere and Oracle are addressing agentic AI security with Oracle AI Database, Deep Data Security, and database-level access controls.
Don’t miss a thing from Oracle AI World 2026. Register for AI World On Air here: https://social.ora.cl/6003BGN9vJ
How can organizations protect sensitive business data as AI agents take on more responsibilities? In this interview on Oracle TV at Oracle AI World 2026, David Knox, vice president of Oracle Database Security Product Management, and Christian M. Grupp, CEO of Sphere, discuss Oracle AI Database security, Deep Data Security, and the importance of protecting information at its source.
Sphere develops AI solutions for complex challenges across multilateral organizations, private sector businesses, and regulated industries. Grupp describes applications involving judicial processes, regulatory monitoring, insurance, and other environments where security requirements are particularly demanding.
A central focus is Oracle's approach to securing data directly within the database. Grupp explains why implementing security at the data layer can simplify policy management and reduce the risks associated with applying different security controls across applications. This approach becomes especially important when AI agents access information on behalf of users and must operate within their assigned permissions.
Knox discusses how agentic AI changes enterprise security requirements. Organizations want to give AI agents access to relevant information, but broader access can introduce unpredictable behavior, security vulnerabilities, and challenges in understanding how automated decisions are made.
The conversation introduces Oracle Deep Data Security and how database-level protections can help organizations manage complex access policies. Grupp discusses how simplifying security implementation can make policies easier to understand, maintain, and apply consistently, reducing the potential for errors in sensitive environments.
The interview also examines Oracle Autonomous Database and its role in supporting secure AI applications that work with multiple types of enterprise data. Both guests emphasize why security architecture must evolve alongside increasingly capable AI agents.
Part of Oracle TV's coverage of Oracle AI World 2026, this discussion explores Oracle AI Database, Deep Data Security, agent permissions, enterprise data protection, and the practical challenges of deploying AI in regulated environments.
Like what you learned about Oracle AI World 2026? Subscribe now https://social.ora.cl/60025k87j
Follow us here:
LinkedIn: https://social.ora.cl/6000B6oR6p
X: https://social.ora.cl/6002B6oRDG
Instagram: https://social.ora.cl/60095eXDN
Facebook: https://social.ora.cl/60065enQp
Securing sensitive enterprise data as AI agents access information and execute business tasks requires consistent permissions, strong database protection, and manageable security policies. Explore how Sphere and Oracle are addressing agentic AI security with Oracle AI Database, Deep Data Security, and database-level access controls.
Don’t miss a thing from Oracle AI World 2026. Register for AI World On Air here: https://social.ora.cl/6003BGN9vJ
How can organizations protect sensitive business data as AI agents take on more responsibilities? In this interview on Oracle TV at Oracle AI World 2026, David Knox, vice president of Oracle Database Security Product Management, and Christian M. Grupp, CEO of Sphere, discuss Oracle AI Database security, Deep Data Security, and the importance of protecting information at its source.
Sphere develops AI solutions for complex challenges across multilateral organizations, private sector businesses, and regulated industries. Grupp describes applications involving judicial processes, regulatory monitoring, insurance, and other environments where security requirements are particularly demanding.
A central focus is Oracle's approach to securing data directly within the database. Grupp explains why implementing security at the data layer can simplify policy management and reduce the risks associated with applying different security controls across applications. This approach becomes especially important when AI agents access information on behalf of users and must operate within their assigned permissions.
Knox discusses how agentic AI changes enterprise security requirements. Organizations want to give AI agents access to relevant information, but broader access can introduce unpredictable behavior, security vulnerabilities, and challenges in understanding how automated decisions are made.
The conversation introduces Oracle Deep Data Security and how database-level protections can help organizations manage complex access policies. Grupp discusses how simplifying security implementation can make policies easier to understand, maintain, and apply consistently, reducing the potential for errors in sensitive environments.
The interview also examines Oracle Autonomous Database and its role in supporting secure AI applications that work with multiple types of enterprise data. Both guests emphasize why security architecture must evolve alongside increasingly capable AI agents.
Part of Oracle TV's coverage of Oracle AI World 2026, this discussion explores Oracle AI Database, Deep Data Security, agent permissions, enterprise data protection, and the practical challenges of deploying AI in regulated environments.
Like what you learned about Oracle AI World 2026? Subscribe now https://social.ora.cl/60025k87j
Follow us here:
LinkedIn: https://social.ora.cl/6000B6oR6p
X: https://social.ora.cl/6002B6oRDG
Instagram: https://social.ora.cl/60095eXDN
Facebook: https://social.ora.cl/60065enQp










