Reco raises $55M to help companies find and control their AI agents as the security market gets crowded

What happened

According to TechCrunch, AI security startup Reco said on Tuesday (September 29, 2026) that it has raised $55 million. The money builds on the $30 million Series B, a later-stage funding round, that Reco raised in February. Customer AT&T invested through its venture arm, alongside Forestay and Quadrille Capital. The new money brings Reco's total funding to $140 million. The company says it will spend it on hiring, sales, partnerships and customer support.

TechCrunch reports that until last year Reco mainly sold software to map and secure SaaS and AI platforms. SaaS means business software used online as a subscription service. Reco has now repositioned around a context graph, a map that links AI agents to the apps, people, accounts and permissions they are tied to. The map lets security teams see what an agent can reach and cut off access it does not need.

Read the full description

Sources are collected, independently checked, then reviewed by Jev.

The "agent sprawl" problem Reco describes

Reco co-founder and CEO Ofer Klein said the biggest change over the past year is that companies are building and deploying AI agents faster than they can keep track of them. He said Reco's platform found 21,000 agents that one of its Fortune 100 customers did not know about. Reco also claims it found an agent at a large financial services customer that had been set up by a former employee. That agent could access Salesforce and share the data with a domain the company could not see.

"The market demand right now for agent security is not only about the agent itself; it's about the entire ecosystem end-to-end," Klein told TechCrunch.

Other companies describe similar problems. According to TechCrunch, Chris Sestito, co-founder and CEO of security startup HiddenLayer, said earlier this month that more than 50 of his customers have AI agents in production touching critical systems and sensitive assets. Another AI startup, Cymphony, said it found about 85,000 files at one US public company that had become accessible to AI tools and agents.

An increasingly crowded market

TechCrunch notes that public Crunchbase and PitchBook profiles show at least two dozen companies selling some form of AI agent security. Their products differ. Their promises to discover and govern agents sound quite similar, though, and involve knowledge graphs, continuous monitoring, runtime security, tool access and MCP vetting.

Reco hopes its existing coverage of SaaS apps will help it stand out. According to TechCrunch, Reco integrates with more than 280 apps, and Klein says new integrations can be added within days. He said the platform uses browser and network signals to find agents outside the apps it connects to directly. It also has controls to inspect prompts, which are the instructions given to an agent, and tool calls, which are the requests an agent makes to other software.

Business figures reported by the company

• Valuation: Klein said it has "more than doubled" since the Series B was announced in February. He estimated it vaguely in the "high hundreds of millions" but would not give specifics.
• Annual recurring revenue (the yearly income from ongoing subscriptions): Klein said it is in the "double-digit millions of dollars" and expects it to triple this year.
• Customers: Reco has more than 100 customers, and financial services companies account for about 40% of the business.

What it means for ordinary readers

AI agents are AI programs that can carry out tasks automatically on a person's behalf and connect to various applications. The situation Reco and others describe may be accurate. If so, companies may be running many unmanaged agents that can reach customer or employee data. One example is an agent set up by a departed employee that can still access business systems. For individuals, this means the safety of personal data depends on more than which systems a company uses. It also depends on whether the company knows which AI programs are touching that data.

If you are employed, learn your company's AI use policy before you connect company systems to AI tools yourself. When you leave or change roles, check whether any automation tools you created need to be handed over or turned off. This article is not a recommendation to buy any product.

Browse the latest news in this topic

• AI news