Operant AI featured in Gartner's new report on How to Secure Custom-Built AI Agents

We are thrilled to share that Operant has been named as a representative vendor in Gartner’s 2025 report on securing custom built AI agents, released in late March. This recognition underscores Operant’s leadership in protecting enterprise grade, autonomous AI systems, particularly as AI agents become central to business workflows.

The Gartner® Report outlines the rise of AI agent autonomous or semi-autonomous systems that use AI to make decisions and take actions and introduces new cybersecurity challenges that must be addressed by businesses. According to the report, 64% of organizations plan to pursue AI agent initiatives within the next year, yet these systems introduce entirely new security challenges that traditional tools can’t handle.

A key insight from the report:

“Through 2029, over 50% of successful cybersecurity attacks against AI agents will exploit access control issues, using direct or indirect prompt injection as an attack vector.”

To address this, Gartner recommends critical security practices including:

• AI agent discovery and monitoring
  
• Stronger, dynamic access control
  
• Tailored development lifecycle and threat modeling
  
• Runtime defenses tailored to AI agent behavior
  

Most importantly, the report emphasizes the need for runtime controls purpose-built for AI, something that traditional application security tools can't fully address.

At Operant AI, we recognize the need to secure AI. Operant’s platform is purpose-built to secure the dynamic and probabilistic nature of AI agents by delivering full visibility, real-time enforcement, and runtime defense. We detect and block threats like prompt injections, jailbreaks, and unauthorized actions through a combination of signature-based and behavioral detection. Our advanced detection capabilities minimizes false positives as Operant intelligently analyzes activities across every layer with our rich first-hand telemetry, enabling security teams to focus on high-priority actions. Beyond detection, we offer robust guardrails and runtime controls that limit undesired agent behavior, even when actions are triggered via APIs or agent memory.

Gartner’s report signals a new phase in AI security—where securing agentic AI becomes mission-critical. At Operant, we’re proud to be helping define this space and provide the tools needed to operate AI safely and responsibly at scale.

We invite you to try Operant’s powerful Runtime AI Protection platform to see for yourself how easy comprehensive security can be for your entire AI application environment.

‍Sign up for a 7-day free trial to experience the power and simplicity of 3D Runtime Defense for yourself.