Blockchain-Enabled Adversarial Threat Intelligence Sharing for Robust Ransomware Detection in Air Gaps

The authors propose a proof of concept (PoC) combining Adversarial Machine Learning (AML) for robust local ransomware detection with blockchain to share adversarial threat intelligence (TI) across connectivity networks (air gaps). Using a ransomware dataset in a simulated air gap environment, they show how independent cloud tenants strengthen local models. Achieved through adversarial threat intelligence training and publishing immutable intelligence on new evasion tactics to a shared ledger, Baseline Machine Learning (ML) models are vulnerable to carefully crafted perturbations. Modern Artificial Intelligence (AI) security adversarial TI training greatly enhances robustness. When a tenant adopts blockchain-anchored intelligence produced by another tenant, it gains significant advantages against previously unseen adversarial ransomware. This PoC demonstrates a collaborative defense approach suitable for critical infrastructure, such as permissioned mission critical service operators, and durable knowledge sharing under strict compliance requirements.