OPSWAT Enhances MetaDefender® Platform with 'MetaDefender Aether' and 'Predictive Alin AI' for Zero-Day Threat Detection

- MetaDefender Aether integrates threat reputation, adaptive sandboxing, AI-powered threat scoring, and similarity-based threat hunting

- Predictive Alin AI enhances pre-execution detection through machine learning-based static analysis

OPSWAT, a global leader in Critical Infrastructure Protection (CIP) cybersecurity solutions, today announced the expansion of its MetaDefender® platform with 'MetaDefender Aether' and 'Predictive Alin AI', now available in Japan. These capabilities enable organizations to detect zero-day and evasive threats earlier, reduce operational overhead, and make more reliable security decisions.

MetaDefender Aether is OPSWAT’s integrated zero-day threat detection solution, designed to inspect files crossing network perimeters—such as file transfers, removable media, email attachments, cloud storage, and web traffic. It combines four layers—'Threat Reputation', 'Dynamic Analysis', 'Machine Learning-based Threat Scoring', and 'AI-powered Threat Hunting'—into a single decision-making process to evaluate threats.

Layer 1 — Threat Reputation (Detection Rate: 48.7%)

Files are evaluated based on OPSWAT’s continuously updated global threat intelligence database. Known malicious files are immediately blocked, while trusted files are fast-tracked, allowing pipeline capacity to be reserved for in-depth analysis when needed.

Layer 2 — Dynamic Analysis (Cumulative Detection Rate: 83.4%)

Files requiring deeper inspection are sent to MetaDefender Aether’s adaptive sandbox. Instead of virtual machines, it uses instruction-level CPU and OS emulation to reveal the full execution path across more than 120 file types. It exposes evasion behaviors commonly used by malware that detects VMs (virtual environments), and newly discovered IOCs (Indicators of Compromise) are fed back into Layer 1 while the file proceeds to downstream AI analysis.

Layer 3 — Machine Learning-based Threat Scoring (Cumulative Detection Rate: 99.3%)

Multiple machine learning engines analyze behavioral signals, anomaly patterns, and IOCs to calculate a risk score that reflects the confidence level of each analysis. This significantly reduces false positives and noise for analysts.

Layer 4 — AI-powered Threat Hunting

Using similarity search, it maps behavioral fingerprints against a database of over 100 million analyzed malware samples, automatically classifying files into known threat families, campaigns, and attack toolkits. Unknown files are transformed into actionable threat intelligence. Based on internal benchmark test data, this architecture achieves up to 99.9% zero-day detection rates and significantly faster analysis speeds compared to traditional sandboxing approaches.

Predictive Alin AI adds a proprietary AI-native layer to the MetaDefender platform. Using machine learning-based static analysis, it evaluates file structure, entropy patterns, and semantic relationships before execution. This enables organizations to respond earlier while minimizing confusion and operational costs caused by false positives. Predictive Alin AI is developed in-house by OPSWAT’s data science and R&D teams and trained on privacy-conscious, curated datasets derived from MetaDefender Aether telemetry, OPSWAT threat intelligence, and investigations by Unit 515. The model is continuously refined through correlated feedback across the entire OPSWAT ecosystem.

"At OPSWAT, we believe prevention starts before execution," said Atsushi Takamatsu, Country Manager at OPSWAT Japan. "With MetaDefender Aether and Predictive Alin AI, we are extending the MetaDefender platform with capabilities that enhance zero-day detection, support more reliable security judgments, and enable faster security decision-making in environments where high accuracy and continuous operation are critical."

MetaDefender Aether and Predictive Alin AI are available for regulated industries such as government, defense, manufacturing, and energy. MetaDefender Aether supports deployment in cloud, hybrid, and air-gapped environments and integrates across the entire MetaDefender ecosystem, including MetaDefender Core, Cloud, Email Security, Managed File Transfer, ICAP, Storage, Kiosk, and Cross-Domain Solutions. Predictive Alin AI is available through MetaDefender Core™ and MetaDefender Cloud™ for Windows and Linux.

For more information, visit MetaDefender Aether and Predictive Alin AI.

About OPSWAT

For over 20 years, OPSWAT has protected the world’s most critical infrastructures across IT, OT, ICS (Industrial Control Systems), cloud, and cross-domain environments. Trusted by government agencies, industrial operators, and Fortune 500 companies worldwide, OPSWAT delivers a prevention-first cybersecurity platform purpose-built for environments where availability, safety, and compliance are non-negotiable.

The OPSWAT MetaDefender platform includes over 20 integrated products that protect diverse segments of critical networks—from endpoint and file security to secure patch management, unidirectional security gateways, cross-domain solutions, secure data transfer, and advanced threat defense. The platform combines Deep CDR™ technology, adaptive sandboxing, Predictive Alin AI analysis, and deterministic data flow control to deliver a unified architecture engineered specifically for OT and regulated environments.

Guided by the principle "Trust no file. Trust no device.™", OPSWAT prevents known, unknown, and AI-generated threats, zero-day attacks, and supply chain risks before they disrupt operations. Through OPSWAT Academy, the company empowers cybersecurity professionals worldwide with training and certification programs.