When companies started deploying AI agents, most treated them like software tools. They got credentials, the same as any other system. A service account here, an API key there. It seemed fine at the time.

It is no longer fine.

According to Okta’s AI Agents at Work 2026 report, 88% of organizations have already experienced suspected or confirmed AI agent security incidents. Ninety percent have no reliable way to govern what their agents are actually doing in production. And yet the average Fortune 500 company is projected to run 150,000 AI agents by 2028, up from fewer than 15 just last year.

That gap, between how fast AI agents are being deployed and how well enterprises can actually govern them, is the market Hush Security is building into.

The Funding

On July 28, Hush Security closed a $30 million Series A round, bringing the company’s total funding to $41 million since its founding in 2024. The round was led by Battery Ventures and YL Ventures, with Akamai Technologies joining as a strategic investor.

The company was founded by Micha Rave, a former Meta Networks executive, and is headquartered in Tel Aviv. Hush emerged from stealth in September 2025 and has been expanding quickly since. In February 2026 they launched what they describe as the first Unified Access Management Platform for agentic AI and non-human identities. In June, they partnered with SDG Corporation to extend security coverage across the broader non-human workforce, which includes service accounts, secrets, and Model Context Protocol servers in addition to AI agents.

The new funding will go toward expanding engineering and sales operations and supporting the growth of their agentic partner ecosystem.

The Problem They Are Solving

Traditional enterprise security was built for humans. You log in, you get access to what you need, you log out. Non-human identities, including AI agents, do not work that way. Most enterprises have given their agents broad, standing credentials that never expire. An agent that needs to read a file gets permanent read access to the entire folder. An agent that needs to send one email gets access to the entire inbox.

This is the core of what security practitioners now call the non-human identity problem. Omdia research found that 96% of organizations are running AI agents on governance models that were not designed for them. When an agent acts unexpectedly, or gets compromised, there is often no registry showing what the agent was, what it had access to, or a way to shut it down.

Hush Security’s platform takes a different approach. Instead of static credentials, agents receive scoped, just-in-time permissions that are automatically provisioned based on the specific task being performed and then revoked when the task completes. The platform includes a central registry of all AI agents running in the environment, audit trails of what each agent accessed and when, a runtime detection engine that surfaces how permissions are actually being used versus how they were granted, and a kill switch to shut down any agent immediately.

The company was selected for the 2026 CrowdStrike/AWS/NVIDIA Cybersecurity Startup Accelerator, which validates the approach is gaining serious traction with enterprise security teams.

What This Means for Business

If you are deploying AI agents in your business, and most businesses either already are or are planning to, this funding round is a signal worth paying attention to.

The pattern playing out right now mirrors what happened with cloud security a decade ago. Companies moved fast to cloud infrastructure, security frameworks lagged behind, and a whole category of security tools had to be built almost from scratch. The same thing is happening with AI agents. The tools to govern them properly are just arriving.

For business owners, the practical takeaway is simple: before you scale your AI agent deployments, make sure you know what each agent has access to, whether those permissions are scoped to what the agent actually needs, and whether you have a way to monitor and shut down agents that behave unexpectedly.

The companies that get ahead of this will deploy AI faster and with more confidence. The ones that ignore it will eventually have a very expensive incident that forces the conversation.

If you want to see how Enterprise DNA approaches building AI agent deployments that are structured and secure from the start, book a discovery call with Sam McKay to talk through your specific situation.