Offroad Inc. launched today with $7 million in funding to build what it calls an agentic identity security team, using artificial intelligence agents to investigate and remediate access risks across human users, machine identities and AI agents.

Founded in September 2025 and based in New York and Tel Aviv, Offroad is targeting a problem that has outgrown manual review. Identity security is no longer simply a question of who has access, the company argues, but a continuous operational challenge of why access exists, whether it remains justified, how it is used and what breaks if it changes. The context needed to answer those questions is scattered across identity providers, human resources systems, software-as-a-service applications, cloud platforms, security tools, tickets and logs.

Rather than surfacing another dashboard of findings, Offroad’s agents gather that context, investigate posture and runtime issues together and then resolve them, either by acting directly where it’s safe or by routing high-impact changes to the right people with the context needed to decide.

The company is pitching AI agents as the reason the old model no longer holds. Autonomous identities now operate at machine speed across many systems at once, breaking the behavioral baselines that runtime detection tools were built on. Service accounts, application programming interface keys and machine identities often carry broad, persistent access with weak ownership and long lifecycles that most organizations lack the processes to govern.

The company put numbers to that risk, auditing 2,890 public OAuth applications across the Google Workspace Marketplace and GitHub Marketplace as of May 2026. It found that roughly one in three, representing more than 1.85 billion installations, showed serious structural security concerns that a careful analyst would likely reject during manual review. Alongside the launch, the company released ohauth.ai, a free OAuth security catalog for reviewing app permissions and governance risks.

“Identity is no longer just a workforce access problem,” explains co-founder and Chief Executive Dan Bendler. “Enterprises now operate across a constantly changing mix of human users, machine identities and AI agents. The context needed to understand and resolve identity risk is spread across dozens of systems and workflows, while security teams are still expected to investigate and remediate issues manually. That model is becoming increasingly difficult to sustain.”

Bendler previously founded two AI startups that together raised more than $45 million. Co-founder and Chief Technology Officer Philip Shteyn is a former captain in Unit 8200, the Israeli military’s elite cyber unit, and helped build the Cortex platform at Palo Alto Networks Inc.

The seed round was led by Ibex Investors and Skywell Capital Partners. Adi Dangot Zukovsky, a partner at Ibex Investors, framed the investment as closing the loop between detection and remediation. “The market is flooded with tools that find identity risk,” she said. “They need operational leverage, not more interfaces.”

Offroad said it will put the funding toward hiring, product development and a bigger U.S. footprint.

Photo: Offroad