Fortinet targets network security platform convergence to address AI-era complexity
SiliconANGLE Thomas Godwin
Fortinet says AI is making security setups messier, so it wants one fabric for networking and security. The pitch is simple: fewer consoles, tighter control, and rules for rogue AI agents.
Based on reporting by SiliconANGLE, Thomas Godwin — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
At Black Hat 2026, Fortinet used Anthony James to make a familiar argument feel newly urgent: the old sprawl of security tools is running out of room in an AI-heavy world. James, Fortinet’s executive vice president of marketing, said the company’s long-running push to unify fragmented products is more relevant now because networking and security are converging anyway.
His core pitch centers on Fortinet Security Fabric. The idea is less about stacking products together and more about building them to work as one from the start. Fortinet wants the data model, the user experience and the policy layer to stay consistent, so operators aren’t stitching together different consoles and trying to reconcile conflicting views when things get messy.
That same logic now stretches into what Fortinet calls the SASE Firewall, which combines secure access service edge and next-generation firewall under the same policy for both on-premises and cloud deployments. James argued that cloud-first SASE vendors pushed too far in one direction, while sovereign AI and agentic workloads have kept the network firmly in the picture.
The newer twist is AI governance inside enterprise environments. Fortinet is building controls for internally deployed AI factories that can decide which agents and MCP servers are allowed, watch token usage and check whether machine-to-machine communications are legitimate. James said customers want visibility into what people are doing with their AI infrastructure, especially when unsanctioned agents or code downloads could push data out to train other models.
My take — AI-written commentary, not fact-checked reporting
Fortinet is betting that AI won’t kill the platform pitch; it’ll finally make it believable. That’s the part everyone keeps learning the hard way: once the tools start talking to each other, the security team wants one set of rules, not a museum of dashboards. Closed, messy stacks look even worse when the machines are the ones moving the data.
Read more about this at: SiliconANGLE