Oracle Pushes Identity Controls as Agentic AI Reshapes Security Operations
Security teams are automating threat response while grappling with how to govern autonomous AI agents. Database vendors like Oracle are redesigning their security tools to track where sensitive data lives and who—or what—can access it.
The way security operations teams manage threats is shifting as they embrace automation to convert raw signals into actionable responses, all while preserving visibility and governance. The challenge lies in calibrating how much decision-making authority to hand to artificial intelligence systems versus retaining human oversight. Companies including Oracle are helping their customers deploy agentic AI safely without exposing databases and sensitive information to new attack vectors.
"Customers are leaning in," said Johnnie Konstantas, group vice president of Oracle Cloud Engineering. "They understand that they have to automate where they can. It's really understanding where the sensitive assets exist and … where in the company AI is being built, particularly agentic AI. That inventory is extremely important."
Konstantas made these remarks during an interview with Dave Vellante at Oracle's "AI Cyberattacks Are Escalating: How to Secure Your Data Now" event, broadcast on theCUBE, SiliconANGLE Media's livestreaming platform. The discussion centered on how identity controls for AI agents are becoming a security priority and how Oracle's tools are evolving to counter AI-driven threats.
Security operations target agentic AI
Industry observers at the Black Hat USA cybersecurity conference have documented a rising emphasis on AI agents within enterprise security discussions. This focus is driving renewed attention to identity management, according to Konstantas.
"Agentic AI brings to the fore this idea that an agent and several sub-agents functioning as a proxy of an individual can access data," he explained. "So, identity takes on a new form in that the agent is kind of carrying those permissions along to access and maybe even take action on a certain workflow. Have good identity in place to the extent that you can have it be uniform and pervasive."
Oracle has incorporated agentic AI capabilities into its security offerings. Konstantas emphasized that the most effective approaches begin with understanding each customer's data landscape and protection priorities.
https://www.youtube.com/embed/Im3BwXoupSk?feature=oembed
"I like to start with customers with what's in it for them rather than the Oracle solution," Konstantas said. "What are you trying to accomplish? Where is your sensitive data? As we look at the data itself, there's a whole spate of capabilities that we've introduced, particularly with the latest version of … the AI database, that are specifically geared toward agentic AI."