Key Takeaways
- Airlock Digital unveiled Agentic AI Control & Governance at Black Hat USA 2026, adding AI‑agent visibility and policy enforcement to its existing preventative endpoint security platform.
- The solution provides command‑ and session‑level insight into trusted AI agents, enabling real‑time governance of what those agents may do on endpoints.
- It combines centralized, version‑controlled policy management for both traditional applications and AI agents with automated discovery and adaptive behavior feedback.
- General availability is slated for Q3 2026, with a preview offered at Booth #5729 during Black Hat USA 2026.
- Industry data shows a pressing need: 82% of organizations have unknown AI agents running, and 65% suffered an AI‑agent‑related security incident in the past year (Cloud Security Alliance, April 2026).
- Airlock Digital positions the new capability as a way to govern AI where work happens—on the endpoint—without hindering business agility.
Overview of the Announcement
Airlock Digital, a recognized leader in preventative endpoint security, introduced Agentic AI Control & Governance at the Black Hat USA 2026 conference in Las Vegas. The announcement marks an expansion of the company’s core application‑control technology to address the rising use of autonomous AI agents within enterprise environments. By delivering visibility into agent behavior and enforcing granular policies at the endpoint, Airlock aims to give organizations the confidence to adopt AI‑driven automation while maintaining strict security oversight.
Why Agentic AI Needs a New Security Layer
Traditional endpoint security solutions focus on deciding what software is allowed to run, but they do not monitor how that software behaves once executed. AI agents, unlike static applications, can evaluate alternatives, adjust their actions, and pursue objectives even when an initial command is blocked. This dynamic behavior creates a governance gap: security teams need to understand ongoing agent activity, define permissible actions, and ensure compliance with corporate policy. Airlock’s new capability directly addresses this gap by adding a second layer of policy that governs what a trusted AI agent is allowed to do after it has been approved to run.
Core Features: Visibility and Discovery
Agentic AI Control & Governance equips security and IT teams with command‑ and session‑level visibility into trusted AI agents. The platform automatically discovers AI applications operating on endpoints and records the specific commands they issue, the sessions they create, and the files they touch. This deep telemetry allows organizations to move beyond simply knowing that an AI agent exists to understanding its actual behavior in real time, a critical step for effective risk management and incident investigation.
Centralized Policy Management
The solution extends Airlock’s existing policy engine to cover both traditional applications and AI agents through a unified, version‑controlled policy repository. Administrators can create, modify, and enforce granular rules that specify permissible actions for each trusted entity, whether it is a legacy executable or an autonomous AI workflow. Changes are tracked, auditable, and can be rolled back, providing the same rigor and compliance readiness that organizations expect from application allowlisting.
Real‑Time Governance and Adaptive Feedback
Airlock’s platform evaluates each AI agent command against the defined policy in real time. When a command violates policy, the system does not merely block the agent; it communicates the decision back to the supported AI agent, allowing the agent to adapt its behavior and continue working within approved boundaries. This feedback loops that mirror how a human employee would adjust to corporate guidelines.
Demonstrating Governance via a Unified Dashboard
To substantiate AI governance, the solution includes a centralized dashboard where security teams can monitor and search AI agent sessions, commands, file interactions, policy decisions, risk indicators, token usage, and associated cost metrics. The dashboard provides a single pane of glass for auditing, reporting, and forensic analysis, helping organizations prove compliance to internal stakeholders and external regulators alike.
Endpoint‑Centric Governance
Unlike cloud‑native AI governance tools that rely solely on platform‑level controls, Airlock enforces its policies where the work actually happens: the endpoint. By governing AI agents at the point of execution—regardless of the underlying AI platform or ecosystem—Airlock delivers consistent, independent policy enforcement across diverse environments. This endpoint‑centric approach ensures that even if an AI agent attempts to bypass higher‑level controls, the endpoint layer still applies the organization’s security intent.
Leadership Perspective on Market Need
David Cottingham, Co‑founder and Chief Product Officer, emphasized that AI agents “don’t simply stop when an action is blocked; they evaluate alternatives and continue working toward their objective.” He argued that instead of repeatedly blocking agents, organizations should define clear operational boundaries that let trusted agents adapt while staying within policy. Kevin Dunne, Chief Executive Officer, echoed this sentiment, noting that customers’ primary challenge is not adopting Agentic AI but governing it effectively. The new capabilities are positioned as a natural extension of Airlock’s leadership in preventative endpoint security, delivering the visibility and control needed to adopt AI without impeding business velocity.
Availability and Demonstration
Airlock Digital plans to make Agentic AI Control & Governance generally available to customers in Q3 2026. A live preview will be offered at Booth #5729 during Black Hat USA 2026 in Las Vegas, allowing security professionals to experience the new features firsthand. Interested parties can also learn more via the company’s blog post titled “AI Agents Behave More Like Employees Than Applications,” which elaborates on the philosophical shift underpinning the technology.
About Airlock Digital
Founded in Australia in 2013, Airlock Digital helps organizations stop threats before they start through preventative endpoint security. The company’s core offerings include precision application control, allowlisting, and operating‑system hardening, enabling customers to run only trusted software and shrink attack surfaces. Airlock serves a broad range of industries—including financial services, healthcare, manufacturing, energy, government, and education—and operates globally from its headquarters in Atlanta, GA. Additional information is available at airlockdigital.com.
Contact Information
For media inquiries, please reach out to Erin Welke, Vice President of Marketing, at [email protected].
This summary captures the essential points of Airlock Digital’s announcement while adhering to the requested length, structure, and stylistic guidelines.