AI agents need their own authentication layer
Scalekit raised $5.5M to build authentication for AI agents. With Gartner predicting 25% of breaches by 2028 will involve compromised agents, traditional identity management needs an upgrade for autonomous workflows.
Scalekit raised $5.5 million to solve a problem most legal teams haven't fully grasped yet: how do you authenticate an AI agent? The San Francisco startup, founded by former Freshworks engineers, builds plug-in authentication protocols for autonomous agents accessing enterprise systems.
Consider the risk profile: Gartner predicts that by 2028, a quarter of enterprise breaches will trace back to compromised AI agents. Meanwhile, 78% of IT executives already worry about controlling these "non-human identities." We're giving agents access to critical systems without the same verification protocols we'd use for human users.
Scalekit uses OAuth 2.0 standards with short-lived, scoped tokens—treating agents like service accounts rather than trying to retrofit human authentication methods. They've landed over 20 customers in three months, which suggests real demand from development teams building agentic workflows.
Product and legal teams face a clear gap in most security frameworks. Current identity management assumes human users who can provide passwords or biometrics. When an agent initiates a database query or API call, traditional authentication breaks down, and so does our ability to maintain audit trails and access controls.
David Prosser
