Fortinet is putting continuous, in-production checks on agentic AI at the center of its security platform. On Aug. 17, 2026, the company announced it had acquired Virtue AI, a firm Fortinet describes as a specialist in AI runtime protection, automated validation, and security for autonomous systems.

Fortinet says the deal extends its existing portfolio, including FortiGate Hyperscale Firewall and FortiAIGate, to cover AI models, applications, and agents from development through runtime. Per the release, Virtue AI’s tooling includes agentic red-teaming across sandboxed environments, discovery of unsanctioned AI apps and agents, MCP tool and source-code scanning, and real-time policy guardrails on text, image, video, audio, and code outputs.

Pre-deployment red-teaming validates a model once, before it ships. Runtime validation targets what changes after deployment: an agent that calls a malicious tool, drifts from policy, or gets manipulated in production. The acquisition prompts CISOs to weigh startup durability and integration risk against platform-native coverage.

Fortinet cites a Gartner forecast that securing AI ecosystems and agents will grow from $2.8 billion in 2026 to $16.4 billion by 2030, a planning input for multi-year AI governance budgets.

Source: Fortinet