Oracle Targets Agentic AI Security Challenges with Database-Level Protections

Oracle Corp. is collaborating with enterprise customers to integrate agentic AI into security operations while addressing emerging challenges in identity and database security. During a recent Oracle event, Johnnie Konstantas, Group Vice President of Oracle Cloud Engineering, outlined the company’s strategy for balancing AI-driven automation with human oversight and robust identity controls.

Oracle Advances Agentic AI Security

Johnnie Konstantas, Group Vice President of Oracle Cloud Engineering, discussed the company’s approach to integrating agentic AI into security operations during a recent Oracle event. According to SiliconANGLE, Oracle is working with customers to automate security workflows while maintaining human oversight and context.

Konstantas emphasized the growing pressure on identity and database security as AI agents become more prevalent in enterprise operations. She noted that identity controls for AI agents must be uniform and pervasive to ensure secure access and workflow execution.

The rise of AI-powered attacks has prompted Oracle to embed security at the database level, moving trust boundaries closer to the data source. This approach aims to reduce reliance on application-layer security and enforce access policies regardless of AI agent instructions.

Database-Level Security and Shared Accountability

Oracle’s latest AI database version includes features specifically designed for agentic AI, according to SiliconANGLE. These capabilities aim to help customers secure data while leveraging AI-driven automation.

The company’s security strategy focuses on shared accountability, a model that divides responsibilities between Oracle and its customers. This shift reflects the complexities of AI-driven incidents, which may exceed the scope of traditional disaster recovery plans.

Konstantas also highlighted the importance of resilience—building security measures into systems before an attack occurs—to address the unique challenges of AI-driven threats.

Industry Focus on AI Security

The event was broadcast exclusively by SiliconANGLE Media and its video platform, theCUBE, as part of their coverage of AI and cybersecurity trends. SiliconANGLE reaches over 15 million tech professionals through its digital ecosystem, while theCUBE Network has over 11,400 alumni in its trusted-based community.

Analysts from theCUBE Research, including Krista Case, participated in the discussion, underscoring the industry’s focus on AI agents and their role in enterprise security.

What this means

Lazyfounder analysis — our interpretation, not reported fact.

For founders and operators, Oracle’s focus on agentic AI security highlights a critical gap in the market: AI-driven automation isn’t just about efficiency—it’s about rethinking security from the ground up.

The shift toward database-level trust boundaries and uniform identity controls for AI agents suggests that startups building AI-powered tools must prioritize security as a core feature, not an afterthought. The emphasis on shared accountability also signals a broader industry trend: vendors may soon be expected to provide not just tools, but frameworks for secure AI deployment.

For B2B startups, this could mean opportunities in developing lightweight, AI-native security solutions that integrate seamlessly with enterprise workflows. For founders in regulated industries (e.g., finance, healthcare), Oracle’s approach underscores the need to align AI adoption with compliance requirements early in the product development cycle.

Key takeaways

• Agentic AI is reshaping cybersecurity, introducing new risks like AI-powered attacks and requiring stronger identity controls.
• Oracle is embedding agentic AI capabilities into its latest database version while emphasizing security at the data source.
• Traditional disaster recovery plans may not suffice for AI-driven incidents, necessitating resilience-focused security strategies.
• A shared accountability model is emerging, redistributing security responsibilities between vendors and customers.
• Human oversight remains critical, even as automation becomes more prevalent in AI-driven security operations.

FAQ

What is agentic AI, and why does it pose security risks?

Agentic AI refers to AI systems that autonomously execute tasks, make decisions, and interact with other systems or users. While this increases efficiency, it also introduces security risks, such as AI-powered attacks, unauthorized access, and vulnerabilities in data handling. These risks require stronger identity controls and database-level protections.

How is Oracle addressing agentic AI security challenges?

Oracle is integrating agentic AI into its security portfolio by embedding security at the database level, enforcing uniform identity controls for AI agents, and promoting a shared accountability model. The company’s latest AI database version includes features tailored for agentic AI to secure data at the source.

What is the shared accountability model in AI security?

The shared accountability model divides security responsibilities between vendors (like Oracle) and customers. Vendors provide tools and frameworks for secure AI deployment, while customers are responsible for implementing these measures and managing their AI-driven workflows.

Related on Lazyfounder

• SiliconANGLE Media
• theCUBE Network
• theCUBE Research
• More ai news
• AWS Launches CloudWatch Omni to Tackle Agentic AI Observability
• Collibra Launches Runtime Governance Tools for Enterprise AI Agents