OPSWAT launches AI file screening engine for MetaDefender

OPSWAT has launched an AI-based pre-execution detection engine for its MetaDefender platform, which is already deployed in customer environments worldwide.

The new product, Predictive Alin AI, is OPSWAT's first proprietary AI-based threat detection engine for MetaDefender. It is designed to assess whether a file is likely to behave maliciously before execution, alongside existing tools in the MetaDefender stack.

The engine uses machine learning-based static analysis to examine file structure, entropy patterns and semantic relationships, rather than relying only on signatures or runtime execution. OPSWAT says it returns verdicts in under 100 milliseconds for most files, uses limited memory and works the same way in online and offline deployments.

OPSWAT positions the software as an added layer for organisations that need to reduce false positives in operationally sensitive settings, particularly in defence, government, manufacturing and energy environments where blocked workflows or unnecessary response actions can disrupt operations.

According to the company, an internal efficacy analysis found that Predictive Alin AI delivered 99.99% precision in identifying safe files across months of production traffic testing. It also reported a measurable efficiency gain when the engine was added to multi-engine deployments.

When the model is uncertain, the MetaDefender platform triggers additional workflows and data handling for further assessment. OPSWAT says these results show the product performs best as a decision-confidence layer within a broader multi-engine architecture.

"At OPSWAT, we've always believed that security begins with prevention, and the assumption that every file is malicious. The Predictive Alin AI Engine wasn't built to replace your security team; it was built to make them more effective and efficient," said Benny Czarny, Founder and Chief Executive Officer, OPSWAT.

"By delivering machine-learning verdicts in milliseconds - before execution, before detonation - we cut through the noise and eliminate the hesitation that costs organizations the most. Our AI-native capabilities give security teams the trust and clarity they need to act with confidence, turning smarter detection into stronger decisions at the speed enterprises demand," Czarny said.

Platform role

Predictive Alin AI joins existing MetaDefender functions including Metascan multiscanning, Deep CDR and adaptive sandbox technologies. The engine was developed in-house by OPSWAT's data science and research and development teams.

The model was trained on what OPSWAT describes as curated, privacy-safe datasets drawn from MetaDefender Aether telemetry, OPSWAT Threat Intelligence and Unit 515 research. The company says it is continuously refined using feedback from across its wider ecosystem.

The launch reflects a broader shift in cyber security towards applying machine learning earlier in the detection process, especially in environments where files cannot easily be detonated or where internet connectivity is limited. In industrial and regulated sectors, tools that can make a rapid judgment before execution may help reduce delays in file handling, though vendors typically combine such models with other inspection methods.

Predictive Alin AI is available through MetaDefender Core for Windows and Linux, as well as through MetaDefender Cloud. OPSWAT says the software performs identically in offline deployments, a feature that may appeal to customers working in segregated or tightly controlled networks.

Yiyi Miao, Chief Product Officer at OPSWAT, said the company prioritised precision over raw detection volume in the engine's design and testing.

"Raw detection rate is not the same as operational value," Miao said.

"Predictive Alin AI was engineered and evaluated with precision as the primary objective. When it fires, customers can have a high degree of confidence in that verdict, which is exactly what many enterprise environments need," she said.

OPSWAT, which focuses on cyber security for critical infrastructure, says MetaDefender now includes more than 20 integrated products covering endpoint and file security, patch management, secure data transfer and cross-domain systems. The company has operated for more than two decades and serves governments, industrial operators and large enterprises. Its latest release is intended to add another screening layer before potentially harmful files reach execution.