"Homeland Security, CISA Builds AI-Based Cybersecurity Analytics Sandbox"

Two of the leading security agencies in the US are building a Machine Learning (ML)-based analytics environment to combat fast-evolving threats and create more robust infrastructures for both the public and private sectors. The Science and Technology Directorate research arm of the Department of Homeland Security (DHS) and the Cybersecurity and Infrastructure Security Agency (CISA) envision a multi-cloud collaborative sandbox that will serve as a training ground for government researchers to test Artificial Intelligence (AI)- and ML-based analytic methods and technologies. It will also incorporate a "loop" of automated ML through which workloads will flow. According to the agencies, the CISA Advanced Analytics Platform for Machine Learning (CAP-M) will facilitate cybersecurity problem-solving for both on-premises and cloud environments. CAP-M will have a multi-cloud environment and multiple data structures, a logical data warehouse to simplify access to all CISA data sets, and a production-like environment to promote realistic testing of vendor solutions. This platform will initially serve cyber missions, but will be adaptable and expandable to accommodate data sets, tools, and collaboration for other infrastructure security missions. The facility will be used for ongoing experimentation in various areas, including data analysis and correlation, to help organizations in adapting to the threat landscape. The data collected from the experiments would be shared with government, academic, and private organizations. The plan involves the protection of privacy and the security of the platform itself. This article continues to discuss plans surrounding the AI-based cybersecurity analytics sandbox.

The Register reports "Homeland Security, CISA Builds AI-Based Cybersecurity Analytics Sandbox"