Fortinet acquires Virtue AI to Expand Security from Networks to AI Agents
- Evelyn Carter
- 5 hours ago
- 5 min read
On August 17, 2026, Fortinet announced the acquisition of Virtue AI, a company specializing in AI runtime protection, automated security testing and autonomous AI system security. Virtue AI will add AI Agent red teaming, monitoring, governance and runtime guardrails to Fortinet’s Security for AI strategy. The event is not associated with a specific CVE.
AI is changing not only how enterprises use software, but also the structure of the attack surface. When an AI Agent can access data, call APIs, connect to tools through Model Context Protocol (MCP), or perform actions autonomously, the security boundary no longer ends at endpoints, applications and traditional networks.
Fortinet’s acquisition of Virtue AI reflects a broader move from protecting network infrastructure toward securing the full AI lifecycle. For Vietnamese enterprises, it is also a reminder that NGFW remains an important network security layer, but it should work together with identity governance, data protection, AI Agent controls and continuous security monitoring.
Why did Fortinet acquire Virtue AI?
Fortinet announced the completion of its acquisition of Virtue AI on August 17, 2026. Virtue AI focuses on three main areas: runtime protection for AI, automated testing and validation of AI systems, and security for autonomous AI environments. Financial terms of the transaction were not disclosed.
According to Fortinet, the acquisition supports its broader Security for AI strategy and its goal of protecting the “agentic enterprise” - an enterprise environment where AI Agents increasingly participate directly in business workflows.
Virtue AI is expected to complement Fortinet’s existing portfolio, including FortiAIGate and other components of the Fortinet AI-Native Security Fabric.
The important shift is the expansion of the security perimeter. Instead of protecting only users, endpoints, workloads, applications and network traffic, enterprise security architectures increasingly need to monitor prompts, models, AI Agents, MCP tools, API calls and AI infrastructure.
What security capabilities does Virtue AI add?
Virtue AI adds four major capability areas to Fortinet’s AI Security portfolio: red teaming for agentic systems, AI Agent governance and monitoring, continuous AI validation, and real-time runtime guardrails.
Capability | Primary Role |
Agentic system red teaming | Tests AI Agents across more than 50 sandbox environments and 14 high-risk domain categories |
Agent protection and governance | Discovers unauthorized AI Agents, inspects MCP tools and source code, and monitors Agent behavior |
Continuous AI validation | Reassesses risk after model updates or policy changes |
Real-time guardrails | Applies security policies to text, images, video, audio and AI-generated code |
Fortinet states that automated red teaming can evaluate hundreds of attack vectors and more than 1,000 risk categories. These scenarios include prompt injection and attacks involving MCP, both of which become more relevant when AI Agents are allowed to access external tools and systems.
FortiAIGate has already been positioned by Fortinet to address risks such as prompt injection, data leakage, model poisoning and excessive resource consumption. Virtue AI broadens this approach with automated testing and runtime monitoring for models, applications and AI Agent systems across development and production environments.
👉 Enterprises looking for more context on how AI Agents expand privilege and attack surfaces can review IPSIP’s analysis of security risks from AI Agents. The key issue is not only what an AI system generates, but also what data, tools and actions the Agent is authorized to access.
Why does this acquisition matter to enterprises?
AI is creating a new category of digital assets that many traditional security governance models do not yet fully track. An AI Agent may use service accounts, call APIs, access databases, execute code or interact with multiple enterprise applications.
As a result, an incorrect or malicious Agent action can have consequences beyond an inaccurate response. It may lead to unauthorized data access, configuration changes, data exposure or other production impacts.
A key consideration for enterprises is understanding the boundary between AI Security and Security Infrastructure. Runtime guardrails can control AI behavior, while Firewall and NGFW platforms control traffic between network zones, applications and services. IAM manages access privileges, while SOC operations correlate and analyze security events across multiple systems.
What should Vietnamese enterprises do about the expanding AI attack surface?
Enterprises do not necessarily need to deploy a complex AI Security platform simply because employees use ChatGPT, Copilot or another AI SaaS product. The first priority is to identify where AI is being used, what data it can access and what actions it is authorized to perform.
Build an inventory of chatbots, AI Agents, AI APIs and AI models used across the organization.
Identify which data each AI system can read, write or transmit externally.
Separate AI Agent identities from normal user accounts where possible.
Apply Least Privilege to APIs, MCP tools, databases and service accounts.
Test for prompt injection, data leakage and tool abuse before deploying AI Agents into production.
Segment AI infrastructure from business-critical systems and sensitive network zones.
Collect logs from Firewalls, endpoints, Cloud platforms, applications and AI systems for monitoring and investigation.
How does IPSIP Vietnam use Fortinet technology for enterprise NGFW environments?
Fortinet’s expansion into AI Security is also relevant to IPSIP Vietnam because Fortinet is one of the technology partners IPSIP works with in enterprise network security projects.
IPSIP’s website identifies IPSIP as a Fortinet partner with experience in consulting, deploying and managing FortiGate solutions for enterprise customers.
Depending on business requirements, IPSIP can assess infrastructure, recommend suitable FortiGate models, design Firewall architecture, implement High Availability and SD-WAN, configure VPNs, develop security policies and support ongoing operational management.
This allows enterprises to build a network security layer based on Fortinet technology rather than simply installing a Firewall appliance and leaving it with a static ruleset.
👉 IIPSIP also provides Managed Firewall 24/7 services, where operational activities can include Firewall monitoring, policy management and optimization, firmware updates, log analysis, configuration change management and support for security alert handling.
👉 Enterprises evaluating different platforms can also review IPSIP’s comparison of six notable enterprise Firewall solutions. IPSIP notes that Fortinet can be well suited to organizations requiring a broad security ecosystem, multiple branch locations or integration with capabilities such as SD-WAN. The final platform choice should still depend on traffic requirements, architecture, security features and operational capabilities.
It is important to distinguish Fortinet’s Virtue AI acquisition from the NGFW systems currently deployed by IPSIP. The acquisition does not mean that Virtue AI capabilities are automatically integrated into every FortiGate appliance or every NGFW environment managed by IPSIP. Fortinet has announced that Virtue AI will strengthen its AI-Native Security Fabric and expand AI runtime protection capabilities.
What does IPSIP Vietnam’s cybersecurity perspective suggest?
Enterprise consequences: If AI systems receive excessive access privileges, an incident can escalate from an incorrect AI output to unauthorized data access, information leakage, inappropriate system actions or increased audit and compliance challenges.
AI Security should not be deployed as an isolated security silo. AI Agents should be incorporated into the same asset inventory, access control, segmentation, logging and incident response frameworks already used for users, applications and workloads.
For Vietnamese enterprises, a layered approach remains the most practical. NGFW and FortiGate can support network segmentation and traffic control; IAM can limit privileges; AI Security technologies can govern models and Agents; and SOC operations can provide continuous monitoring and incident response.
References
Fortinet - Fortinet Advances Continuous AI Protection with the Acquisition of Virtue AI: https://www.fortinet.com/corporate/about-us/newsroom/press-releases/2026/fortinet-advances-continuous-ai-protection-with-the-acquisition-of-virtue-ai
Fortinet Investor Relations - Fortinet Advances Continuous AI Protection with the Acquisition of Virtue AI: https://investor.fortinet.com/news-releases/news-release-details/fortinet-advances-continuous-ai-protection-acquisition-virtue-ai/
Comments