OPSWAT has announced that its Deep CDR technology achieved a 100% sanitization rate in independent testing by AV-Comparatives. The result came from more than 300 test cases spanning roughly 50 file formats.

This marks the third time an independent lab has reported a 100% protection score for Deep CDR. SE Labs and SecureIQLab previously recorded similar results. Notably, Deep CDR was also the first Content Disarm and Reconstruction (CDR) technology to score 100% accuracy in SE Labs testing.

AV-Comparatives is an Austrian, ISO 9001:2015-certified testing lab. For this evaluation, it assessed Deep CDR on sanitization effectiveness by attack scenario, coverage across around 50 file types, and file fidelity after reconstruction. The test set included AI-assisted malicious content, zero-day exploits, malicious macros, embedded executables, and nested archives, among other evasive file-borne threats. Deep CDR passed every case under the tested configuration, while also preserving file usability.

“Macros, embedded executables, nested archives, AI-assisted payloads, 300-plus cases, and nothing came through,” said Thomas Uhlemann, Cybersecurity Evangelist at AV-Comparatives. “Our test is pass or fail. There is no partial credit.”

Attack campaigns continue to evolve quickly. However, the underlying exploitation techniques, such as malicious macros, embedded executables, steganography, and malformed archives, remain common tools for attackers. Deep CDR takes a prevention-first approach to file security, treating every file as untrusted by default. It then disarms and rebuilds each file in milliseconds, removing embedded and out-of-policy content. At the same time, it preserves layout, formatting, tables, and formulas, so the file remains usable. Deep CDR supports more than 200 file types across email, web, file-transfer, and endpoint workflows. As part of the MetaDefender platform, it can be deployed on-premises, in the cloud, or in air-gapped environments.

“At OPSWAT, we do not grade our own homework,” said Benny Czarny, Founder and CEO of OPSWAT. “Our commitment to CDR means repeatedly putting Deep CDR Technology in front of independent security testing organizations and asking them to stress test it against the newest file-borne attack techniques, including threats created or adapted with AI.”

He added that three independent labs have now reported 100% protection results for the technology. According to Czarny, that level of evidence reflects what critical infrastructure requires. Detection methods must first recognize a threat before they can stop it, he noted. Deep CDR, by contrast, removes risky content regardless of whether the threat is known, unknown, or AI-generated.

Related post

View all