Archives

HiddenLayer Announces Machine Learning Security Collaboration with Intel

HiddenLayer

HiddenLayer, a Gartner recognized AI Application Security company, is proud to announce a strategic collaboration with Intel Corporation to deliver a comprehensive hardware and software-based ML security solution for organizations in regulated and compliance-focused industries.

Recognizing the current gap in ML scanning within a secure environment, Intel and HiddenLayer have come together to offer customers an end-to-end ML model protection platform. This technology collaboration leverages the scale, performance, and privacy of Confidential Computing, built with HiddenLayer’s cutting-edge adversarial ML threat protection capabilities and Intel SGX.

Customers can now access an encrypted ML Model Scanner designed to protect their ML models. Intel SGX enables the smallest trust boundary for Confidential Computing. It’s the most researched, updated, and deployed confidential computing technology in the data center on the market today. This is critical for driving the continued adoption of AI across industries.HiddenLayer, with support from Intel, is committed to building one of the industry’s most secure ML model integrity platforms. By running the HiddenLayer Model Scanner in a Trusted Execution Environment (TEE), customers can help safeguard their most valuable intellectual property: their AI.

Also Read: Alteryx Ventures Announces Strategic Investment in Fiddler to Boost Machine Learning Operations for Customers

The HiddenLayer ML Model Scanner delivers:

  • Malware Analysis: Scans ML models for embedded malicious code that could serve as an infection vector and launchpad for malware.
  • Vulnerability Assessment: Scans for known CVEs and zero-day vulnerabilities targeting ML models.
  • Model Integrity: Analysis of ML model’s layers, components, and tensors to detect tampering or corruption.
  • Comprehensive Detection: Utilizes a combination of static detection, dynamic analysis, and machine learning techniques to identify malware, vulnerabilities, model integrity, and corruption issues.
  • Catalog a Known-Good State: Baseline your ML models for identifying future tampering.

Broad Support: Supports various ML model file types, including Pickle, Dill, Joblib, Numpy, Zip, ONNX, and HDF5.

Intel’s commitment to cybersecurity is evident in its extensive Confidential Computing portfolio. Confidential Computing powered by Intel SGX offers a TEE to help protect sensitive data.

Industries with strict regulatory and compliance requirements can significantly benefit from Confidential Computing. Intel SGX is continuously improved over time, with deployments across major cloud service providers including IBM, Alibaba, Baidu, and Microsoft.

SOURCE: PR Newswire