CIQ launched RLC Pro Hardened and Ascender Pro deployment for US federal agencies, providing real-time kernel exploit detection and automated remediation within CISA's Binding Operational Directive 26-04 compliance framework. The directive requires federal systems to remediate highest-risk vulnerabilities within three days of appearing in the Known Exploited Vulnerabilities catalogue. RLC Pro Hardened is the first Enterprise Linux distribution shipping with runtime kernel exploitation detection enabled by default through Linux Kernel Runtime Guard. The system validates kernel integrity continuously and records exploitation as it occurs, before patches become available. Ascender Pro adds Reaqt, an event-driven engine that monitors fleet logs and automatically executes remediation playbooks. The combined deployment addresses compliance gaps when exploits emerge before patches are released. Agency remediation policies must support BOD 26-04 by 7 August 2026.

CIQ, the founding commercial sponsor of Rocky Linux, has announced new features for its Ascender Pro automation platform. The platform now automatically detects and fixes problems across Enterprise Linux fleets without human intervention. New additions include Ascender Reaqt, which watches inventory and fixes issues automatically, and Ascender Registry, which gives teams control over collections and containers. Federated Inventories speeds up job routing to the correct inventory. CIQ has also open sourced Galaxy Proxy, a caching layer that reduces wait times for Ansible Galaxy responses by 75 to 80 percent. The tool is available to the wider Ansible community, whilst Ascender Pro customers receive the full Registry application. Ascender Reaqt, Ascender Registry and Federated Inventories are now available to existing Ascender Pro customers.

CIQ has announced that its Fuzzball platform now delivers a production-ready AI compute and inference environment for NVIDIA DGX Spark. The platform enables AI teams to develop, tune and deploy workloads on a single DGX Spark system, then scale to thousands of GPUs without rebuilding. Fuzzball eliminates months of infrastructure assembly by providing ready-to-run environments with built-in workflow templates. The platform supports sovereign AI deployments where data cannot leave premises, allowing private model tuning whilst maintaining infrastructure control. Multiple DGX Spark systems can work together under a common Fuzzball environment, with seamless scaling to larger NVIDIA GPU deployments. The platform bridges local DGX Spark systems with HPC clusters and cloud capacity under a single operational model, addressing the traditional split between AI experimentation and production deployment.