March 2026 Summaries

Modern enterprise architecture faces the challenge of managing three distinct traffic patterns—external APIs, internal microservices, and AI/LLM calls—each with unique demands but often handled by separate tools, leading to gateway sprawl, fragmented security, and scattered observability. The global API management market is expected to grow significantly by 2033, making a unified control plane, which centralizes policy enforcement, observability, and governance across all traffic types, increasingly crucial. Kong offers a platform that addresses these challenges by providing a single control plane that integrates North-South API traffic, East-West microservices, and AI/LLM traffic management under one consistent framework, eliminating the inefficiencies of managing multiple tools. This approach not only reduces operational overhead and improves security and observability but also enhances strategic agility by allowing organizations to switch providers and manage costs efficiently. As digital transformation accelerates, adopting such a unified platform can turn architectural complexity into a competitive advantage, fostering innovation and streamlining compliance across diverse environments.

AI APIs play a critical role in modern applications, driving costs and potential revenue while facing unique challenges in operations, security, and governance. Unlike traditional APIs, AI APIs, especially those using large language models, are unpredictable and can lead to significant infrastructure costs due to their volatile nature. Monetization strategies must go beyond pricing models to include robust infrastructure controls such as rate control, quota enforcement, consumer segmentation, and usage visibility. The API gateway becomes essential as the enforcement layer where these controls are applied, ensuring that AI services are protected and monetization efforts are effective. The integration of governance and monetization at the gateway level is crucial, as it allows for real-time enforcement of policies, prevents abuse, and aligns infrastructure costs with revenue models. This approach is particularly important as AI systems become more autonomous, requiring well-structured control mechanisms to prevent unintended cost spikes and maintain sustainable operations.

Enterprise teams assessing API testing tools such as Insomnia and Postman must consider the cost implications, particularly as Postman has shifted previously free collaboration features behind a paywall, starting at $19 per person per month. This change means even small teams face increased costs solely to maintain API definitions alongside their code. In contrast, Insomnia offers free Git sync for small teams, allowing real-time collaboration without additional fees. Insomnia's Pro plan, which is more cost-effective than Postman's, includes role-based access controls and security features without extra charges, highlighting the company's commitment to affordability and transparency. Postman's introduction of AI features, while innovative, has been used as a justification for increased pricing on basic functionalities that were once complimentary, representing a shift towards premium pricing tiers. Meanwhile, Insomnia's forthcoming integration with Kong Konnect aims to streamline API management by providing developers with direct access to a central repository, enhancing workflow efficiency. The fundamental difference lies in Postman's focus on visibility and Insomnia's emphasis on active integration, ensuring developers have a working source of truth.

Event-driven architectures (EDAs) have become integral to modern enterprises, particularly with Apache Kafka leading the transition from synchronous REST APIs to asynchronous event streams to support real-time analytics and AI. However, the security of these systems has been a significant challenge, especially in aligning modern web identity standards with native event stream protocols. The release of Kong Event Gateway v1.1 addresses this issue by introducing OAuth Token Claim Mapping for Policies and native mTLS Authentication, bridging the security gap between traditional API security models and Kafka's infrastructure-centric posture. This update allows organizations to enforce fine-grained, identity-aware access control across event streams, supporting more secure multi-tenant environments while maintaining alignment with existing Kafka security practices. By integrating OAuth-based policies and mTLS configurations, Kong Event Gateway v1.1 enables enterprises to scale their event-driven architectures securely, ensuring that security and identity are at the forefront rather than an afterthought.

Kong assures its customers that it is unaffected by the recently publicized PyPI LiteLLM incident, as it does not utilize LiteLLM in its runtime stack. The LiteLLM versions 1.82.7 and 1.82.8, distributed via PyPI, contained a malicious script capable of stealing credentials from environments where it was installed. While Kong's products are not impacted, organizations using LiteLLM independently should treat any environment that used pip install for these versions as potentially compromised. Other popular projects, such as CrewAI and Camel-AI, might also be affected if installations occurred during a specific four-hour window. For further inquiries about Kong's security or AI infrastructure, customers are encouraged to contact their Kong account team or the security team via email.

Kong has appointed Bruce Felt as its Chief Financial Officer amidst a transformative phase in enterprise product management, driven by AI tools that enhance productivity by 35% and offer a significant time-to-market advantage, according to Gartner research. The report highlights companies like Kong, Zapier, and Leena.ai for their innovative use of AI in product development, with Kong's SVP of Products, Reza Shafii, illustrating how AI-generated prompts streamline their workflow by feeding directly into prototyping tools and allowing a fail-fast approach. This strategy reduces rework costs and accelerates product delivery, reflecting Kong's commitment to adopting AI internally and offering a platform that manages both traditional API traffic and AI workloads. Kong emphasizes its dynamic use of AI across its organization to enhance speed and efficiency, inviting potential partners to experience the benefits of its AI-driven infrastructure firsthand.

Kong has introduced a plugin for Solace upstream mediation that allows users to bridge the gap between REST APIs and event brokers like Solace Broker, providing a unified interface for handling both APIs and events. This integration enables developers to expose Solace event streams through the Kong Gateway while applying consistent policies and developer experiences used for REST APIs. By running Kong in front of a Solace Broker, users benefit from enhanced authentication, access control, schema validation, data transformation, and observability, ensuring secure and efficient event processing. The setup involves a series of configurations using Docker, Solace Cloud, and Kong's declarative configuration tool, decK, to create a seamless mediation between HTTP requests and Solace events. This integration supports advanced features like rate limiting, request size caps, and dynamic topic routing, making it easier to manage both APIs and events with enterprise-level governance and protection.

AI connectivity has emerged as a critical infrastructure for managing the rapid and complex growth of AI interactions within enterprises, addressing operational chaos caused by the proliferation of AI models and providers. As organizations move beyond initial AI adoption, they face challenges such as fragmentation, Shadow AI, escalating token costs, and security vulnerabilities, necessitating a centralized control plane like the AI Gateway. This gateway facilitates a unified strategy integrating APIs, events, and context engineering, optimizing AI interactions by ensuring secure, efficient, and cost-effective operations. AI connectivity extends traditional API management by incorporating AI-specific capabilities such as semantic caching, multi-model routing, and specialized security measures. The convergence of API and AI management is vital as enterprises employ multi-LLM strategies to maintain flexibility and competitiveness, requiring platforms like Kong AI Gateway to provide centralized governance, reduce costs, and enhance reliability through comprehensive monitoring and consistent policy enforcement.

Kong has introduced Bruce Felt as its Chief Financial Officer while unveiling the general availability of Managed Redis cache for Kong Dedicated Cloud Gateways (DCGW), which aims to streamline multicloud deployments and enhance the governance of AI operations. The Managed Redis cache provides a high-performance, shared state layer that integrates seamlessly with Kong Konnect, enabling enterprises to manage AI token usage, agent communication, and performance optimization without needing complex infrastructure. This feature transforms DCGW into a state-aware platform, offering capabilities like AI quota governance, semantic caching, and context management, which are crucial for managing AI models and ensuring consistent performance across multicloud environments. By eliminating the need for manual infrastructure management, the service simplifies the deployment process, allowing for consistent global policies across cloud providers like AWS and Azure. Kong's expansion into platforms like Google Cloud, alongside its compliance with SOC 2 and PCI DSS standards and a 99.99% SLA, reflects its commitment to providing secure, reliable, and scalable API infrastructure, essential in an era of increasing AI integration and architectural modernization.

Kong has introduced a Managed Redis cache add-on for its Dedicated Cloud Gateways (DCGW) to address the challenge of state synchronization in multicloud architectures and AI systems. This add-on is now available on AWS and Azure, providing enterprises with a high-performance, fully managed shared state layer within Kong Konnect. It simplifies governance and enhances performance across multicloud deployments by integrating the speed of an in-memory data store with the ease of a SaaS product, all without the need for infrastructure management. Enterprises can deploy gateways across different cloud providers while maintaining centralized governance through Kong Konnect, enabling intelligent, stateful operations such as caching and AI governance. The architecture involves a SaaS control plane, cloud-hosted gateway data planes, and optional regional shared states powered by Redis, allowing for consistent global configuration and local traffic processing. This setup supports various use cases, including API response caching, rate limiting, and AI token management, and can be automated using infrastructure tools like Terraform, facilitating scalable and resilient multicloud architectures.

In the evolving AI token economy of 2026, enterprises face significant financial challenges as output tokens—generated by AI models in response to input prompts—are considerably more costly, often costing 3 to 10 times more than input tokens due to their resource-intensive, sequential processing nature. This disparity presents a risk of cost overruns, especially with Agentic AI systems that engage in multi-step reasoning and recursive loops. To mitigate these risks, Kong offers solutions such as the Kong AI Gateway, which enforces directional guardrails and token-aware rate limiting, and Konnect Metering & Billing, which enables precise monetization of AI usage through flexible rate cards. These tools provide critical financial oversight, allowing organizations to control AI costs efficiently and transform token consumption into profitable revenue streams. Moreover, the AI token pricing landscape, which had been on a downward trend due to increased model efficiency and competition, is experiencing upward pressure due to infrastructure constraints and the introduction of next-generation models with premium capabilities. As businesses navigate this complex environment, fluency in token economics and the implementation of robust cost governance frameworks become essential for maintaining profitability and leveraging the full potential of AI-driven technologies.

Kong has introduced Bruce Felt as its new Chief Financial Officer amidst the growing adoption of AI-assisted coding tools like Claude Code, developed by Anthropic, which is revolutionizing software engineering by allowing autonomous editing and execution of code from natural language instructions. Despite its potential to enhance productivity and redefine engineering workflows, the widespread deployment of Claude Code raises significant governance challenges, including security risks, cost management, and compliance issues. Without centralized control, organizations risk uncontrolled expenses, data leakage, and auditability problems. Kong proposes its AI Gateway as a solution to these challenges by providing a centralized control mechanism that manages authentication, rate limits, logging, and secure model routing. This infrastructure is increasingly seen as essential for responsibly scaling AI applications, providing organizations with critical governance capabilities without disrupting developer workflows. As AI coding tools continue to integrate deeply into development processes, the role of such gateways in managing the associated risks becomes pivotal in maintaining security and compliance standards.

Kong highlights the shift towards usage-based monetization as a pivotal development in the digital platform industry, emphasizing the need for flexible and programmable billing systems. This approach, demonstrated by Kong Konnect Metering & Billing, addresses the limitations of relying on a single payment provider like Stripe, which can become a bottleneck as businesses grow and diversify their offerings. By decoupling usage metering from payment execution, Kong offers the ability to customize billing solutions to suit different customer needs, from self-service subscriptions to complex enterprise contracts. This flexibility allows companies to optimize margins by routing billing through various channels, such as keeping credit card charges on platforms like Stripe for smaller accounts while using internal ERP systems for large enterprise deals to minimize processing fees. Kong's solution ensures financial infrastructure supports business scalability and adaptability, providing the architectural freedom to adjust pricing models and partnerships without being locked into a specific payment provider.

Metered billing for APIs involves charging customers based on actual usage, such as the number of API calls or data processed, rather than a flat fee, aligning costs with the value delivered. This model supports transparency and fairness by ensuring users pay only for what they use, which can boost customer loyalty and optimize revenue as customer growth directly correlates with spending. The architecture for effective metered billing requires a robust system with layers including usage events, meters and aggregation, rating and price models, and invoicing and settlement, each crucial for accurate measurement and financial scrutiny. Key components like idempotency prevent double-charging, while an event-driven architecture ensures replayable, auditable, and billing-grade telemetry. This approach not only enhances trust and reduces revenue leakage but also supports hybrid pricing models, which are increasingly favored in the market for their flexibility and potential for higher growth rates. Tools like Kong Konnect and OpenMeter facilitate the transition to usage-based billing, providing automated reconciliation and a seamless "Ingress-to-Invoice" solution that turns API traffic into reliable revenue streams.

Kong's Insomnia has been recognized in the Gartner Market Guide for API and MCP Testing Tools, affirming its status as a robust platform for modern API development and testing. As one of only eight vendors offering native MCP server testing, Insomnia enables engineering teams to confidently build and validate agentic applications using the broader Kong ecosystem. The platform supports security and contract testing, providing flexibility in data storage and collaboration controls, which is crucial for enterprises managing sensitive API schemas. Gartner highlights Insomnia's AI-powered capabilities, which streamline workflows and enhance test coverage, and its integration with Konnect, enabling seamless discovery and testing of APIs and MCP servers. As the demand for MCP-related inquiries grows, Insomnia is poised to meet the evolving needs of engineering teams by expanding its AI endpoint support, security policies, and integration capabilities, ensuring readiness for the shift to agentic architecture.