Sophos has announced cybersecurity datasets, tools and methodologies that leverage artificial intelligence to help organizations combat cyberattacks, according to a prepared statement.
Sophos, backed by private equity firm Thoma Bravo, unveiled the following AI cybersecurity offerings:
1. SOREL-20M: This is a malware research dataset with a curated and labeled set of samples and security-relevant metadata, Sophos noted. It was developed by SophosAI data scientists and threat intelligence provider ReversingLabs and contains metadata, labels and features for 20 million Windows Portable Executable (WPE) files and 10 million disarmed malware samples.
2. Impersonation Protection Method: SophosAI's Impersonation Protection method is designed to protect organizations against spear-phishing attacks, Sophos said. With this method, organizations can compare the display name of inbound emails against high-level executive titles and flag suspicious messages.
3. Epidemiology-Based Statistical Models: SophosAI has created epidemiology-based statistical models for estimating the prevalence of malware infections, the company indicated. These models can help organizations identify malicious "dark matter," malware that might be missed or wrongly classified, and "future malware."
4. YaraML Automatic Signature Generation Tools: SophosAI has developed YaraML, an automatic signature generation method that allows organizations to compile machine learning models into signature languages, the company stated. YaraML lets organizations use AI to "write" signatures to detect malware.
Along with the aforementioned cybersecurity offerings, Sophos provides AI-powered and cloud-native solutions that organizations can use to secure their endpoints and networks against ransomware, malware and other cyber threats. Sophos also offers Rapid Response and other cybersecurity solutions that MSPs and MSSPs can integrate into their portfolios.
