Skip to main content
From the Asset Scans page, you can click a Model Artifact to view the Detection Summary, Attributes, and related MITRE ATLAS techniques.
Detection CategoriesScan results include a Detection Category and Threat Level to provide some insight into risks related to the model file. See Detection Categories for more information.
  1. In the Console, select Supply Chain > Asset Scans in the side navigation.
    • Detections by Threat Level: Bar chart for detections, colored by threat level.
      • Bar chart displays the last seven days
      • Displays the total number of scans
      • Threat levels included are: None, Low, Medium, High, and Critical
    • Asset Compliance: Bar chart for compliant and non-compliant assets (based on scan data).
      • Bar chart displays the last seven days
      • Displays the total number of assets scanned
      • Compliance data included are: compliant and non-compliant
    Asset Scan Expand Row
  2. Click the green arrow next to the scan name to expand the row and view more information. See Column Descriptions for information about data in each table column.
    Asset Scan Expand Row
  3. Click a tab to display specific model scans.
    • Detections - Displays model scans with detections discovered.
    • Compliant - Displays model scans that are compliant with the assigned policy.
    • Noncompliant - Displays model scans that are not compliant with the assigned policy.
    • All Scans - Displays all model scans.
    Model Scanner page
  4. Click the green arrow next to Threat Level to view the Detection Summary page that contains model scan information.
    • Scan summary: Displays key scan information, including the related policy and compliance status.
    • Scan Results: Displays the types of files analyzed, along with the number and percentage of files that have security issues.
    Detection Summary Section
  5. The Categories tab includes Detections and All Categories related to the scan results. See Detection Categories for more information.
    All Categories
  6. The Attributes tab includes a list of files related to the scan and details about the selected file.
    Attributes
  7. The MITRE ATLAS tab highlights MITRE ATLAS techniques related to the scan.
    • Click Show Full Attack Chain to see all MITRE ATLAS techniques.
    MITRE ATLAS

​
Column Descriptions

Column Description
ScanThe name of the model scanned.
VersionThe version ID for the scan.
Detections / FilesThe number of severities found over the total number of files discovered in the scan.
AdvisoriesKnown files of concern, but are not exploits in and of themselves. Advisories should be reviewed prior to model usage.
Scanned OnThe date and time of the scan.
Scan Policy StatusThe status displays Compliant or Noncompliant based on the policy assigned. Unknown usually means no policy is assigned.
RegionDisplays the region where the files are located, if available.
ProviderDisplays the cloud provider, if available.
Scanner VersionSupply Chain scanner version used to scan the model.

Origin

Where the scanned model was found and how the scan was started.

This is displayed as [Platform or Catalog] | [Initiation Method].

  • Platform or Catalog: Where the scanned model was found.

    • Sources identified by HiddenLayer include:

      • Unknown Origin
      • Hugging Face
      • Local File System
      • databricks
      • S3.

  • Initiation Method: How the scan job was started in HiddenLayer.

    • HiddenLayer initiation methods include:

      • UI Upload: Scan started from the Console, using a local file upload or community scan.
      • API Upload: Scan started from the HiddenLayer API.
      • Integration: Scan started automatically by a configured integration.
      • Unknown Source: Initiation method could not be determined.

  • Example combinations:

    • Hugging Face | UI Upload - Hugging Face model scanned by the Console community scan; manually initiated by a user.
    • Databricks Unity Catalog | Integration - Databricks artifact scanned via a HiddenLayer integration.
Actions
  • Model Intelligence: Delivers critical insights into model architecture, ancestry, and task alignment to support trust, security, and compliance in AI deployment. See Model Intelligence for more information.
  • AIBOM: Provides an inventory of AI artifacts for every scanned model in an industry-standard SBOM format, empowering security and data science teams to track AI components, vulnerabilities, and ensure compliance with regulatory standards.
Threat LevelThe severity level of the scan.