Infisical raises $40 million to secure enterprise AI agent access

Securing the AI workforce

Infisical, a cybersecurity company based in San Francisco, has successfully raised $40 million in a Series B funding round. This investment brings the company's valuation to $600 million and comes at a time when businesses are rapidly integrating artificial intelligence into their daily operations. The funding was led by investor Elad Gil, who also participated in the company's previous Series A round.

The company plans to use this new capital to accelerate its product development and grow its engineering and go-to-market teams. They also intend to strengthen their security infrastructure as more organizations begin to deploy AI agents across their internal systems. Other notable investors in this round include Gradient and Y Combinator.

According to reports from citybiz, the goal is to address the specific cybersecurity challenges created by autonomous systems. As these agents begin to perform more complex tasks, they require deeper access to a company's private data and infrastructure. Infisical wants to ensure that this increased functionality does not come at the cost of corporate security.

How Agent Vault works

A major part of Infisical’s expansion involves a product called Agent Vault. This tool is designed to let AI agents interact with enterprise systems without ever receiving sensitive credentials like passwords, API keys, or authentication tokens. By removing the need for the agent to "know" the password, the company reduces the risk of these secrets being exposed through cyberattacks or external services.

The technology allows human administrators to define exactly which systems an AI agent is permitted to access. This creates a controlled environment where the agent can perform its designated job while the underlying credentials remain hidden. This approach mirrors how humans are managed in a company, where employees are rarely given unrestricted access to every single system.

Vlad Matsiiako, the Co-Founder and CEO of Infisical, explained the logic behind this approach. He noted that "AI agents shouldn’t be any different" than human employees regarding access. The company's platform is built to protect critical systems such as databases, cloud environments, and source code repositories used by both humans and automated processes.

Market adoption and scale

Infisical is already a significant player in the credential management space, serving more than 1,300 customers and 300,000 developers worldwide. Their platform is currently responsible for securing approximately 10 billion secrets on a daily basis. This scale demonstrates a high level of trust from major global brands that rely on the company's infrastructure.

The company's customer list includes prominent names such as Airbus, Volkswagen, Hugging Face, OpenRouter, Lucid, First American Financial Corporation, Wise, and Clay. These organizations use Infisical to manage digital certificates, encryption keys, and privileged access to their most sensitive infrastructure. This diverse client base shows that the need for secure credential management spans across various industries.

Beyond its enterprise reach, Infisical has also seen significant interest from the developer community. Its open-source project has garnered more than 30,000 stars on GitHub. This level of developer adoption highlights the importance of its technology in the broader software development lifecycle and its role in securing the modern tech stack.

The next decade of security

The investment comes as the industry shifts from AI applications that simply generate content to agents capable of executing actual tasks. These new agents need to interact with operational systems and access databases to be useful. This transition introduces complex requirements for authentication and access controls that traditional security models may not be equipped to handle.

Investor Elad Gil highlighted that autonomous AI agents are becoming a new class of users within every enterprise. He suggested that managing the permissions for these systems will be a primary security hurdle for the next ten years. Because these agents operate across multiple applications, managing their permissions becomes a moving target for security teams.

Infisical's latest funding will specifically support the development of tools meant to address these new requirements. By focusing on AI-driven workflows, the company aims to help enterprises adopt these technologies safely. The goal is to provide a framework where automation can flourish without creating new vulnerabilities in the company's core infrastructure.