Keyfactor Debuts 'PKI for AI' to Secure Autonomous Systems on AWS

CLEVELAND, OH – December 16, 2025 – Digital trust leader Keyfactor has launched its re-architected Command MCP Server in the Amazon Web Services (AWS) Marketplace, a move that signals a foundational step toward securing the next wave of artificial intelligence. The offering aims to establish what the company calls “PKI for AI,” a framework for providing unique, verifiable digital identities to autonomous AI systems, tackling a critical and emerging security challenge before it becomes a systemic vulnerability.

This release expands on an earlier prototype, introducing a scalable, multi-user server designed to support a wide array of AI clients. By making this experimental deployment accessible through AWS, Keyfactor is inviting developers and security professionals to explore how to build trust into the very fabric of an increasingly autonomous digital world.

The Rise of Agentic AI and the Trust Deficit

The industry is rapidly moving beyond generative AI that simply responds to prompts. The new frontier is 'Agentic AI'—advanced systems capable of setting their own goals, formulating multi-step plans, and executing complex tasks with minimal human intervention. From automating supply chains to managing complex IT infrastructure, these autonomous agents promise unprecedented efficiency. However, their very autonomy creates a profound security dilemma.

Without a robust identity framework, these agents represent a massive, uncharted attack surface. Traditional security measures like static API keys are ill-suited for dynamic, autonomous entities that can reason and act at machine speed. Industry analysts have warned that treating these agents like human users with password-based credentials is a recipe for disaster. This concern was underscored by a widely discussed simulation of a state-sponsored attack in September 2025, which demonstrated how an AI-driven campaign could perform reconnaissance, harvest credentials, and move laterally across networks autonomously, overwhelming conventional defenses.

This new threat landscape demands a paradigm shift from human-centric security to machine-centric trust. Experts suggest that the solution lies in providing AI agents with their own workload-bound identities, such as cryptographic certificates, that are intrinsically tied to their function and cannot be easily stolen or replicated. This approach is central to building a Zero Trust architecture that can withstand the speed and scale of AI-driven operations.

Building a Foundation: 'PKI for AI' Explained

Keyfactor's answer to this challenge is 'PKI for AI,' a concept that applies the time-tested principles of Public Key Infrastructure to the world of artificial intelligence. The core idea is to issue a unique, non-replicable X.509 certificate to every AI agent, effectively giving each one a cryptographically verifiable identity. This digital birth certificate allows an organization to authenticate, authorize, and audit the actions of every AI agent operating within its environment.

This framework enables several critical security functions. It ensures that communication between AI agents, or between an agent and a data source, is secured with mutual TLS (mTLS), preventing eavesdropping and man-in-the-middle attacks. It provides a mechanism for ensuring the integrity of AI-generated content and decisions through digital signatures. Most importantly, it establishes clear accountability, extending Zero Trust principles to non-human actors and preventing them from becoming security weak points.

The Command MCP Server functions as the bridge between AI models and Keyfactor's certificate lifecycle management platform. It utilizes the emerging Model Context Protocol (MCP), which enables AI agents to securely interact with enterprise systems and tools. “This new architecture is a major milestone in connecting AI systems with digital trust infrastructure,” said Ted Shorter, CTO of Keyfactor. “By making it available as an experimental deployment on AWS, we’re inviting the community to explore how AI agents can accelerate zero trust deployments and other PKI initiatives, and how those same systems can, in turn, secure the AI ecosystem.”

Streamlining Adoption Through the AWS Cloud

Making this technology accessible is as important as the innovation itself. By listing the Command MCP Server in the AWS Marketplace, Keyfactor is significantly lowering the barrier to entry for developers, researchers, and organizations looking to experiment with AI security. The offering is delivered as a ready-to-run container image, simplifying deployment and testing within secure AWS environments.

The server's architecture represents a significant evolution from its initial desktop-only, single-user prototype. The new remote server architecture is designed for scalability, allowing multiple AI clients—whether on web, desktop, or mobile—to connect securely to a single, centrally managed instance. This provides unified visibility and oversight of the deployed infrastructure, a critical feature for enterprise-grade control.

This launch also highlights Keyfactor's deepening strategic collaboration with AWS. The company is part of the AWS Independent Software Vendor (ISV) Accelerate Program, a co-sell initiative that aligns its solutions with the AWS global sales force. This partnership aims to help organizations scale their PKI and machine identity management in the cloud, positioning Keyfactor as a key enabler of digital trust for multi-cloud enterprises and DevOps workflows.

A Strategic Play in an Emerging Market

With the market for AI in cybersecurity projected to grow from around $22 billion in 2023 to over $60 billion by 2028, Keyfactor's move is a strategic play to establish a foothold in a critical and nascent segment. While numerous companies offer AI-powered security analytics or model protection, Keyfactor is carving out a unique niche by focusing on the foundational identity of the AI agents themselves.

Company leadership is clear that the Command MCP Server is an early-stage, experimental release not yet intended for production use. Community engagement is still in its infancy, with its AWS Marketplace listing yet to receive reviews and its corresponding GitHub repository showing minimal activity. This transparency underscores the pioneering nature of the project; Keyfactor is not just launching a product, but seeding an ecosystem and shaping a standard for a problem that is just beginning to be fully understood.

By proactively addressing the security of Agentic AI, the company is positioning itself as a thought leader and an essential infrastructure provider for the next generation of digital transformation. This initiative aligns with its broader mission to help organizations manage the growing complexity of machine identities, automate certificate lifecycles, and prepare for future cryptographic challenges like the transition to post-quantum cryptography. While the vision of a fully trusted, autonomous AI ecosystem is still on the horizon, laying the architectural groundwork for that trust begins now.