
C1 is an identity security company that governs access for employees, non-human identities (NHIs) and AI agents from one platform. Security and IT teams use it to decide who and what can reach applications, infrastructure, models and tools, to grant that access only for as long as it is needed, and to prove afterwards that it was right.
The platform grew out of identity governance: access requests, access reviews and automated provisioning across connected applications. During 2026 C1 extended the same identity and policy foundation to AI, with products for AI tool provisioning, agent identity, MCP and LLM traffic, credential vending and a vault.
At A Glance
| Vendor | C1 (C1.ai) |
|---|---|
| Category | Identity governance and administration, just-in-time access, NHI and AI agent access governance |
| Identities covered | Human users, service accounts, workloads, applications, AI tools, AI agents and MCP servers |
| Deployment | SaaS hosted on AWS; C1 AppHub is self-hosted |
| Agentic AI | AI Access Management, C1 Agents, MCP Gateway, LLM Gateway, shadow AI discovery, credential vending |
| Certifications | SOC 2 Type II (annual audit) |
| Customers | Instacart, Zscaler, DoorDash, Qualtrics, Ramp |
| Funding | Series B (2025) |
What C1 Does
Access in most organisations is still granted through IT tickets and left in place long after the work is done, and AI agents add a fast-growing population of identities that need keys, tools and data. C1 replaces that with policy. People and agents ask for access, policy decides whether it is approved automatically or sent to a human, the grant is scoped and time-bound, and every step is logged.
Everything sits on one identity graph and policy engine. Governance products handle the lifecycle of human and machine access. The AI products apply the same rules at runtime, sitting between agents and the models, tools and credentials they use. A connector library links the platform to the systems where access is actually granted and removed, and a programmable layer lets other systems and agents build on C1 directly.
Products
Identity Governance
Teams that still grant access through tickets often struggle to prove at audit time that it was right. The governance core replaces that with a managed access lifecycle. Users request access through Slack, the C1 web app or other channels, and approval policies route each request. Low-risk requests can be approved automatically, while sensitive ones go to an approver.
- Just-in-time, time-bound grants that expire automatically, reducing standing privilege
- User access reviews and certifications that confirm or revoke access for compliance programmes
- Automatic provisioning and deprovisioning across connected applications
- Import of access data from airgapped applications
NHI And Identity Security
Many security teams cannot say how many service accounts and agents they have, or who owns them. This product builds an identity inventory across people, service accounts, other NHIs and AI agents and finds risky conditions in real time, such as unowned service accounts, stale access and agents with no owner. Each finding becomes a remediation that runs through C1 workflows or the tools the security team already uses, so every fix is governed and auditable.
AI Access Management
Announced in March 2026, AI Access Management brings governance to AI tools, personal assistants and enterprise agents. It keeps the rollout of AI tools under control instead of leaving it to shadow adoption. Employees request an AI tool and are provisioned within seconds, through auto-approval or a routed approval. Agents are treated as first-class identities with their own credentials, policies, lifecycle states and owners, and the MCP connections they use to reach enterprise systems come under the same visibility and policy.
C1 Vault
C1 Vault gives people and workloads a safer place for secrets than shared files or environment variables. It encrypts secrets on the member’s device before upload, so C1 stores only ciphertext and only members’ devices hold the keys that can read it. Workloads exchange a signed workload identity for a scoped credential instead of holding the underlying key, and access changes trigger immediate revocation and re-keying.
Credential Vending And C1 Egress
Applications often hold long-lived keys that nobody tracks. Credential vending, launched on 30 September 2026, issues every credential an application needs, each tied to one role, limited to approved IP ranges, time-bound and logged. An inventory shows who created each credential, what it can reach and when it was last used. C1 Egress goes a step further and swaps in the real secret at the network layer, so the application never holds it. Requests to internal addresses and cloud metadata endpoints are blocked by default.
C1 AppHub
C1 AppHub addresses the internal apps teams now build quickly, often with AI coding agents, that end up outside access reviews. Launched on 28 September 2026, it is a self-hosted platform that customers run in their own cloud to deploy, authenticate and set permissions for the applications their teams build. It ships as a public GitHub repository. Each app is registered in the identity graph so reviews and offboarding cover it.
Headless Identity Infrastructure
Launched in May 2026, Headless Identity Infrastructure is a programmable layer that exposes C1’s identity data, policy and workflows through APIs. It suits platform teams that want to embed access requests and policy checks in their own tools, and agents that need to act on governance without working only through the C1 interface.
How It Works
| Component | What it does |
|---|---|
| SaaS platform | C1 runs on AWS across multiple Availability Zones and holds identity data, policy, workflows and the audit trail for human, non-human and agent identities in one system. |
| Connectors | A library of pre-built connectors reads and changes access in each target system for discovery, provisioning, revocation and reviews. Custom connectors are built with the open-source Baton SDK, and configuration-driven options exist for REST APIs, SCIM and SQL. |
| Gateways | The MCP Gateway and LLM Gateway sit in the path of agent traffic and apply per-agent scopes and runtime policy, using the same identity and policy foundation as the governance products. |
| Data protection | Traffic to C1 uses TLS 1.2 or higher and internal services use mutual TLS. API keys and secrets are encrypted with AWS KMS symmetric keys and encrypted again at rest, and internet-facing API services cannot decrypt data. |
Identities And Systems Covered
| Human users | Employees and contractors requesting, reviewing and holding access |
|---|---|
| Non-human identities | Service accounts, workloads, integrations and applications in the identity graph |
| AI identities | AI tools, personal assistants, enterprise agents, coding agents and MCP servers |
| Credentials | API keys and secrets held in C1 Vault or issued through credential vending |
Agentic AI Capabilities
C1 Agents And C1 Autonomous Worker
C1 Agents are C1’s own AI agents for identity work such as compliance scanning and access management. C1 Autonomous Worker (C1AW), launched in June 2026, is the agent customers interact with day to day, and it reaches the control plane through C1’s APIs and MCP servers. Each job runs in an isolated workspace with only the identity and permissions assigned to it. Sensitive changes go to a person for review, and every action lands in the audit trail like any other access event.
MCP Gateway And C1 MCP Servers
The MCP Gateway governs MCP, tool and API actions made by agents, scoping each agent to the tools its job needs and checking calls against policy at runtime. C1 MCP Servers expose C1 to agents, and an AppHub MCP server publishes apps built by AI coding agents.
LLM Gateway
Announced on 1 October 2026, the LLM Gateway routes inference requests across public, private and customer-controlled model deployments.
Shadow AI Discovery
Shadow AI discovery finds unsanctioned AI tools, agents, MCP servers and credentials so they can be inventoried and given a business owner and a control owner.
Integrations
| Area | Integrations |
|---|---|
| Identity and HR | Okta, Microsoft Entra ID, Google Workspace, Workday |
| Cloud and data | AWS, Google Cloud Platform, Kubernetes, Snowflake |
| Applications | GitHub, Bitbucket, Jira, Salesforce, ServiceNow, Oracle NetSuite, PagerDuty |
| Infrastructure as code | Terraform, for access configuration |
Company Background
| Founders and leadership | Alex Bovee (CEO and Cofounder), Paul Querna (CTO and Cofounder), Will Bengtson (CISO) |
|---|---|
| Offices | San Francisco and Portland, United States |
| Funding | A $79 million Series B in October 2025 led by Greycroft, with CrowdStrike Falcon Fund, Accel and Felicis, taking total funding past $100 million |
| Acquisition | Workday announced a definitive agreement to acquire C1.ai on 13 October 2026, expected to close in the fourth quarter of Workday’s fiscal year 2027 |
| Recognition | Included in the SACR Runtime Security for AI Agents report (2026), the Citizens JMP Cyber 66 (2026) and CRN’s 10 Cybersecurity Startups to Watch (2026) |
Note: This overview is compiled from C1’s publicly available information as of 3 November 2026. It is provided for information only. NHI Mgmt Group does not endorse any vendor, and nothing on this page represents the opinions of NHI Mgmt Group.