As the industry races toward autonomous networks, Kilter by BackBox takes a deliberate approach, keeping NetSecOps in command of every decision.
BackBox, a network cyber resilience company, announced Kilter AI, the new AI-powered intelligence layer of its Network Cyber Resilience Platform, now named Kilter. Kilter AI arrives with a distinct perspective on where AI belongs in network management. While many in the industry are eager to hand over critical infrastructure to fully autonomous systems, BackBox developed Kilter AI based on a different principle: AI should serve as a trusted advisor rather than a robot on autopilot. This approach ensures that network operations and security teams (NetSecOps) maintain control.
Kilter AI is designed to mirror the BackBox principles for responsible AI, introduced in Oct. 2025. These principles emphasize trustworthiness, responsibility, safety, and security while delivering consistent, accurate results that improve over time. It keeps humans at the center of the decision-making process, treating AI as a trusted advisor rather than the sole decision-maker. Additionally, it is transparent about how it generates recommendations and the data it uses, ensuring that every stakeholder can understand and trust its output.
“BackBox enables customers to explore agentic NetSecOps at their own speed and understand how they want to use AI-powered solutions to manage their network infrastructure,” said Rekha Shenoy, CEO of BackBox. “The built-in AI intelligence in Kilter AI leverages BackBox’s sizable network infrastructure knowledge and automation library, providing new insights, recommendations, and automations that significantly enhance security and reduce manual work.”
Why a Human-in-the-Loop Approach Matters
The stakes for getting AI right have never been higher. More than 48,000 new CVEs were published in 2025, a 20% increase from the previous year, and about 60% of breach victims cite a known, unpatched vulnerability. Mythos has given attackers the ability to exploit weaknesses in days instead of months or years. For NetSecOps teams managing too many tools across diverse environments with limited time and staff, the pressure to automate is intense, but handing over the network to a robot that no one can question only trades one risk for another.
Kilter AI is built to responsibly ease that pressure by accelerating work while keeping expert judgment involved. The demand is clear: 60% of IT leaders plan to implement AI-enabled predictive network automation within the next two years.
“Today’s diverse network environments are becoming more complex,” said Alexander Neubauer, Head of Network & Security, ACP IT Solutions GmbH. “BackBox stands out as an essential solution for collecting information on all our devices and allowing us to efficiently track inventory in a single centralized location. When new devices are added, tasks are automated to ensure compliance with our configuration policy.”
Also Read: CIO Influence Interview with John Elliott, Cybersecurity Author Fellow at Pluralsight
Putting the Approach into Practice
Kilter AI provides a connected set of capabilities that guide teams from raw vulnerability data to actionable remediation, with human teams reviewing and approving before anything is implemented. These include:
- Enriched Data Aggregation: a single, normalized source of truth for network vulnerabilities that combines contextual CVE severity data from CISA, NVD, NIST, and vendor sources for accurate risk data on each device, allowing teams to quickly prioritize the greatest risks.
- CVE Workaround Insights: normalized workaround data with clear sources and links to vendor details to understand if a config change mitigates a CVE, reducing the need to patch or upgrade; plus, the ability to search configuration files to verify if action is needed on all affected devices.
- CVE Workaround to Chains: AI-curated workaround data transformed into intuitive, visual, executable chains that teams can review and adjust before running to automate CVE remediation, to streamline workflows, and reduce errors.
- Automation Creation Assistance: a unified interface that turns CLI input into production-ready, device-aware automations that teams can preview, save, execute, or reuse as backups, tasks, or IntelliChecks. Device knowledge is enough to get you started; there is no need to be a scripting expert.
One Platform, Three Layers
Kilter AI is the intelligence layer of the Kilter Platform, a single no-code foundation that unifies three layers for network resilience. Kilter ALM automates the device lifecycle — onboarding, backup and recovery, upgrades, and patching — across more than 180 vendors with 3,000+ pre-built automations. Kilter Enterprise handles compliance, policy, and network infrastructure integrity, from a configuration source of truth and hardening frameworks to vulnerability intelligence and change monitoring.
Where Kilter AI Is Headed
Kilter AI forms the foundation of BackBox’s plan for agentic NetSecOps: networks that can reason, adapt, and act increasingly independently. BackBox will grow that autonomy one deliberate step at a time, building trust with each new capability so teams can delegate more tasks only as they feel comfortable.
BackBox is already earning early validation for its approach to AI, including recent recognition by the 2026 A.I. Awards for Best AI-driven Automation Solution and from Cyber Defense Magazine in the AI/ML for Cyber Resilience.
Catch more CIO Insights: How Are CIOs Aligning Technology with Workforce Agility?
[To share your insights with us, please write to psen@itechseries.com ]
