Reco Raises $55 Million as AI Agent Security Becomes the Next Big Enterprise Battle

The rapid adoption of AI agents is creating a new security problem for businesses: companies are increasingly struggling to know exactly which agents are operating inside their networks, what they can access and where their data can go.

Reco is betting on that problem with a fresh $55 million funding round as the market for AI agent security becomes increasingly crowded.

The funding extension builds on the company's $30 million Series B announced in February and brings Reco's total capital raised to $140 million. AT&T, through its venture arm, joined Forestay and Quadrille Capital in the latest investment.

Reco was initially focused on helping companies map and secure SaaS and AI platforms. As businesses began deploying AI agents at a much faster pace, the startup expanded its platform to track the broader relationships between agents, applications, employees, accounts and permissions.

That shift reflects a growing challenge for enterprise security teams. According to Reco CEO and co-founder Ofer Klein, one Fortune 100 customer discovered 21,000 AI agents that its security team was previously unaware of.

At another financial-services customer, Reco says it identified an AI agent created by a former employee that could access Salesforce data and share it with an external domain that the company could not see.

The company is building what it calls a context graph to give security teams a broader view of how AI agents interact with enterprise systems. Its platform can identify agents, map their access and provide controls for managing what those agents are allowed to reach.

Reco says its system currently integrates with more than 280 applications, while browser and network signals can also help identify agents operating outside directly connected applications. The platform additionally provides controls for inspecting prompts and tool calls.

Reco is entering an increasingly competitive market. TechCrunch identified at least two dozen companies working on different aspects of AI agent security, ranging from monitoring the tools agents use to controlling their access to sensitive data and detecting unauthorized AI deployments.

The broader concern is tied to the speed at which businesses are moving AI agents into production. Security companies are increasingly warning that once autonomous systems begin interacting with critical enterprise infrastructure, sensitive information and business applications, the potential exposure can expand rapidly.

Reco says it now has more than 100 customers, with financial-services companies accounting for roughly 40% of its business. The company reported annual recurring revenue in the double-digit millions of dollars and expects that figure to triple this year.

The startup's valuation has more than doubled since its February Series B, according to Klein, although the company has not disclosed a specific figure.

Reco plans to use the new capital to expand hiring, sales, partnerships and customer support.

The latest funding underscores a broader shift in enterprise cybersecurity. As AI agents move from experimental tools to systems capable of accessing applications, data and corporate infrastructure, securing the agents themselves is becoming a new category of enterprise security spending.