Keycard Launches Identity and Access Solution for AI Agents

Keycard has launched a new identity and access platform specifically designed for AI agents, enabling organizations to manage and secure automated systems with the same level of control as human users. This platform integrates directly with existing corporate identity solutions, allowing businesses to confidently deploy AI into production environments. It identifies each AI agent, assigns permissions based on specific tasks, and dynamically enforces security policies while maintaining a complete audit trail of all activities. The system ensures that AI agents can only perform the precise actions intended by their human creators and users.

Ian Livingstone, CEO of Keycard, emphasized the platform’s importance, stating, “AI agents represent a once-in-a-generation shift, greater than the SaaS and cloud wave combined. But without trusted access controls, they can’t leave the lab. Keycard provides the guardrails that allow agents to act safely on behalf of people and businesses, unlocking the true potential of the agent economy.”

The company was founded by Ian Livingstone, Matthew Creager, and Jared Hanson, who bring extensive experience from the security and identity sectors. Livingstone and Creager previously held senior roles at Snyk, where they helped scale platform engineering as the company’s revenue grew substantially. Hanson was the Chief Architect at Auth0, later joining Okta, and is the creator of the widely-used Passport.js authentication framework for Node.js.

During their careers, the founders repeatedly observed developers struggling to securely connect various services and applications at an enterprise level. These integration challenges frequently resulted in security breaches and significant project delays. The rise of AI agents has intensified this problem. These autonomous systems now operate by the thousands across different platforms and organizational boundaries, requiring dynamic permissions that change based on the specific task, the involved parties, and the resources being accessed.

Traditional identity solutions were designed for human users performing manual, point-and-click operations, making them inadequate for the dynamic, ephemeral, and autonomous nature of AI agents. According to industry analysis, AI is projected to generate the largest number of new identities with privileged access, highlighting the urgent need for specialized security solutions.

Keycard’s platform addresses this gap with several core innovations. The system provides cryptographic proof of an agent’s identity and authorization using federated, standards-based protocols, ensuring compatibility with leading AI platforms from companies like Anthropic, Microsoft, and OpenAI while avoiding vendor lock-in.

The platform replaces vulnerable static API keys with dynamic, identity-bound tokens that are scoped to specific tasks. These tokens link directly to the agent’s verified workload identity, computing provider, or hardware device, establishing end-to-end trust without requiring code modifications. This approach allows security policies to automatically adapt to evolving trust relationships between users, systems, and organizations.

Contextual access controls are enforced in real-time during operation. Through Keycard’s identity-aware data model, users can define policies based on relationships and tasks, with the ability to instantly revoke access through a simple API call when needed.

Built for internet-scale performance, the infrastructure supports the dynamic, temporary, and performance-intensive nature of AI agents. It models the complex relationships between human and machine identities along with their associated resources, representing a fundamental architectural departure from conventional solutions.

For developers, Keycard offers software development kits that simplify building trusted agent applications without requiring deep security expertise. This provides security teams with necessary context and feedback mechanisms to govern AI systems during development rather than after deployment.

The company actively contributes to emerging standards including Model Context Protocol (MCP), WIMSE, and OAuth extensions for agents. Keycard has implemented the first production-ready support for OAuth 2.1 Client ID Metadata Documents within the MCP framework.

In a significant funding announcement, Keycard revealed it has secured $38 million through seed and Series A financing rounds. The $8 million seed round was co-led by Andreessen Horowitz and boldstart ventures, while Acrew Capital led the $30 million Series A. Notable angel investors participated, including executives from Groq, Chainguard, Datadog, and former leaders from Okta and Auth0.

Zane Lackey, partner at Andreessen Horowitz, commented, “This is the Auth0 moment for agent access. The agent ecosystem needs foundational authorization infrastructure. This team has the rare combination of infrastructure expertise and standards leadership required to build it.”

Ed Sim, general partner at boldstart ventures, added, “Winning in agentic security takes a rare mix: build for developers and nail security from day one. Ian, Matthew and Jared did it before at Snyk and Auth0 and they’re doing it again with Keycard. This is the team to define the identity and trust category for the agent era.”

Asad Khaliq, founding partner at Acrew Capital, concluded, “Trusted agents define the next chapter of computing. Developers will lead the way and need durable identity and access foundations. Keycard extends trust and control to the agent layer.”

The newly acquired capital will support continued advancement of Keycard’s identity platform and expansion of its research and development team.

(Source: HelpNet Security)