Systems | Development | Analytics | API | Testing

In the rapidly evolving landscape of AI-driven development, the Model Context Protocol (MCP) has emerged as the critical standard for connecting AI applications to the data and tools they need. We are excited to announce the Technical Preview (TP) of Kong MCP Registry, a major milestone in our mission to provide the most comprehensive platform for modern API and AI management.

While every organization races to deploy AI agents faster, a quieter crisis is compounding in the background, and it will play a large part in determining who survives the agentic era. The numbers are stark. Too many executives see AI governance as a brake on innovation or something to figure out later, after the speed problem is solved. With agentic AI, that's backwards.

While every organization races to deploy AI agents faster, finance departments are watching something alarming unfold—and it will play a large part in determining who survives the agentic era. The numbers are stark: 84% of companies report more than 6% gross margin erosion from AI costs. Within that, 26% report erosion of 16% or more. And only 15% of companies can forecast AI costs within ±10% accuracy—the majority miss by 11-25%, and nearly one in four miss by more than 50%.

Modern AI applications are no longer just about sending prompts to an LLM and returning text. As soon as AI systems need to interact with real business data, internal APIs, or operational workflows, the problem becomes one of orchestration, security, and control. The challenge is to build secure AI agents without embedding fragile logic or exposing sensitive systems directly to a model. This is where a layered architecture using Volcano SDK, DataKit, and Kong MCP Proxy becomes compelling.

Have you ever asked an AI assistant to generate code for a framework it doesn't quite understand? Maybe it produces something that looks right, but the syntax is slightly off, or it uses deprecated patterns. The AI is working hard, but it lacks the specific context it needs to truly help you. The Model Context Protocol (MCP) was designed to bridge this knowledge gap by giving AI assistants access to domain-specific knowledge and capabilities they don't have built in.

See how easily Konnect Metering & Billing transforms API and AI traffic management into new revenue streams. We've talked about why 2026 is the year of AI unit economics. There, we explored the "2025 hangover" where organizations realized that without financial governance, AI isn't just a science project but has become a margin-bleeding cost center. But "governance" and "monetization" shouldn't just be buzzwords in a resolution; they need to be part of your active infrastructure.

Today, we're excited to announce Kong Mesh 2.13. Kong Mesh 2.13 delivers full support for Mesh Identity for Kubernetes and Universal mode. Plus, it's been designated as a Long Term Support release, with support for a total of 2 years. But first, what's Kong Mesh for the uninitiated?

KAi, the AI assistant inside Kong Konnect, just got significantly more capable. Today, we're announcing an enhanced beta version powered by the new Kong Konnect MCP Server — a shared infrastructure layer that also opens up your API platform to IDE copilots and custom agents. The result? KAi can now do things it couldn't before, and those same capabilities are available wherever you work. If you've used KAi before, you'll notice the difference immediately.

AI agents are spreading across organizations rapidly. Each agent needs secure access to different Model Context Protocol (MCP) servers. Authentication becomes complex. Scaling creates bottlenecks. The dreaded "too many endpoints" problem emerges. You face a classic AI infrastructure headache. The numbers tell the story. Organizations using AI in at least one business function jumped from 55% to 78% in just one year. Generative AI usage specifically rose from 33% in 2023 to 71% in 2024.

Gartner just published research that should be required reading for every platform and infrastructure leader building for the agentic era. The report, "How to Enable Agentic AI via API-Based Integration," makes a stark claim: incrementally reworking existing APIs and connector-based integrations for AI agents is no longer sufficient.