Fortinet has announced the launch of its FortiGuard Inline AI Security Service, a cloud-delivered offering that extends AI-specific threat inspection and security guardrails to FortiGate customers. The new service enables organizations to monitor AI activity, prevent emerging AI threats, and reduce sensitive data exposure using their existing Fortinet security infrastructure.
The announcement builds on ongoing advancements in FortiAIGate, which provides AI security controls, enhanced visibility into shadow AI and autonomous agents, and expanded compliance and access management capabilities. The platform also incorporates technology from Fortinet’s recent acquisition of Virtue AI, strengthening its ability to deliver high-performance AI guardrails.
By leveraging Fortinet’s extensive FortiGate deployment base, the company aims to simplify enterprise adoption of AI security. Organizations can activate the FortiGuard Inline AI Security Service without installing or maintaining additional dedicated AI security infrastructure. For enterprises with stricter privacy, security, and data sovereignty requirements, FortiAIGate remains available for deployment within their own environments, providing greater control over AI traffic processing and data handling.
“Organizations need AI security that can scale with how quickly AI is being adopted across the enterprise,” said Michael Xie, Founder, President, and Chief Technology Officer at Fortinet. “By extending advanced AI inspection and guardrails through FortiGate, we are making AI security available through the infrastructure customers already use. FortiAIGate provides the foundation for this broader strategy, giving customers flexibility in how they deploy and consume AI security based on their cost, performance, privacy, and sovereignty requirements.”
The FortiGuard Inline AI Security Service introduces deeper AI traffic inspection directly through existing FortiGate infrastructure. Powered by FortiAIGate-based guardrails, the service evaluates prompts and exchanged data in real time, applying security, compliance, and data protection policies to interactions involving AI agents, large language models (LLMs), and Model Context Protocol (MCP) tools.
The offering complements Fortinet’s existing capabilities for detecting shadow AI activity and enforcing data loss prevention (DLP) policies. Rather than limiting protection to identifying unauthorized AI services or monitoring sensitive data transfers, the new service examines the content of AI interactions and can mask confidential information before it is exposed. This expanded protection is delivered without requiring customers to deploy separate security infrastructure.
As enterprises increasingly integrate LLMs and autonomous AI agents into production environments, these technologies are gaining access to confidential information, enterprise applications, and business-critical processes. Conventional cybersecurity controls typically examine IP addresses, signatures, and network payloads, but they are not designed to interpret the security implications of AI prompts and generated responses.
This creates additional attack surfaces where malicious instructions embedded in prompts, model outputs, or MCP communications can influence AI behavior. Such attacks may result in unauthorized tool execution, harmful API requests, or unintended actions by autonomous agents.
FortiAIGate addresses these challenges by establishing a real-time security control point between AI applications and models. The platform evaluates interactions and applies policies according to the AI model, application, user privileges, and relevant business context.
Its expanded capabilities include:
- Real-time AI threat prevention and data protection: FortiAIGate applies multilayered security guardrails to prompts, model outputs, MCP tool requests and responses, API interactions, and autonomous agent activities. Inspection supports text, images, audio, video, and code while maintaining low latency and minimizing false positives. Enhanced DLP capabilities identify 35 categories of personally identifiable information (PII) and confidential data, enabling organizations to generate alerts, block requests, or redact sensitive content before disclosure.
- AI usage governance and cost management: Administrators can configure model access permissions, spending budgets, consumption thresholds, and rate limits at the organization, workspace, project, or individual user level. Notifications are generated as spending approaches configured limits, while requests exceeding those thresholds can be blocked. The platform also prevents malicious prompts and Denial-of-Wallet attacks from consuming tokens or GPU resources. Intelligent routing selects appropriate AI models based on request content, security policies, and operational costs without requiring application re-architecture.
- Centralized governance across business units: Dedicated workspaces allow departments and teams to maintain separate AI models, policies, activity logs, roles, and permissions while remaining subject to centralized security visibility and administrative oversight.
- Compliance enforcement and audit readiness: Organizations can implement more than 50 predefined policies covering AI governance standards, data privacy legislation, industry regulations, and cybersecurity frameworks. These policies can be customized to specific operational requirements, with detailed audit logs documenting prompts, generated outputs, and tool interactions.
- Visibility into autonomous AI agents: FortiAIGate provides insight into AI agents operating throughout the enterprise, including their connections to business applications, APIs, tools, datasets, and MCP servers. This visibility helps security teams understand how agents interact with enterprise resources.
- Flexible deployment and data sovereignty: Organizations can deploy FortiAIGate within on-premises infrastructure or private cloud environments using Kubernetes and NVIDIA GPUs. Self-hosted deployments keep prompts, model responses, sensitive data, and operational logs within the organization’s controlled environment.
Through its combination of cloud-delivered and self-hosted deployment models, Fortinet is seeking to consolidate AI security capabilities while reducing the operational burden associated with managing multiple disconnected security products.
“FortiAIGate takes an elegant approach to a complex problem,” said Will Townsend, Chief Analyst at LoneStar Advisory & Research. “Rather than asking customers to manage separate tools for AI security, governance, cost, and agent visibility, Fortinet is bringing these capabilities together in a single gateway. The rapid integration of Virtue AI technology further strengthens that story by improving the guardrails and performance customers need as they move AI into production.”
The FortiGuard Inline AI Security Service and FortiAIGate extend Fortinet’s broader security architecture by integrating AI inspection, runtime protection, and governance with capabilities spanning network security, application security, cloud security, data protection, identity, and security operations.
Together, these offerings provide organizations with a more consistent framework for addressing AI-related risks across users, applications, models, autonomous agents, and underlying infrastructure. The approach is designed to support enterprise AI adoption while accommodating performance expectations, privacy obligations, regulatory compliance, and data sovereignty requirements.
Join our LinkedIn group Information Security Community!
