Glasswall has introduced Foresight, an AI-based threat prediction product that assesses the likelihood of files being malicious by analyzing structural data generated during its Content Disarm and Reconstruction process, reports Security Brief New Zealand.
Integrated into the Glasswall Meteor file-cleaning application, Foresight transforms sanitization output into threat intelligence signals for investigation and triage. The tool produces probabilistic classifications for files, including zero-day threats, based on large-scale analysis of indicators drawn from millions of samples. Unlike traditional approaches relying on signature scanners, sandboxing, or external threat feeds, Foresight derives signals from deterministic file-structure analysis, enabling operation in offline and air-gapped environments common in national security and critical infrastructure. Security teams can use contextual risk scores to prioritize investigations and feed structured data into SIEM and SOC workflows.
Glasswall reports a false positive rate of 0.015% for PDFs and similarly low rates for common formats. Chief Product Officer Paul Farrington noted that combining CDR assurance with risk insight provides a "clearer understanding of the hostile file activity entering environments," particularly where conventional sandboxing falls short.
