Notebook instances and connected data pipelines may be configured with unrestricted egress paths, such as file downloads, public network interfaces, or writable external destinations. Training datasets, model artifacts, and embedded credentials could be transferred out of the environment through these paths. This results in a loss of confidentiality for proprietary data and models, and may expose regulated or sensitive records used in training.
Training Data or Model Artifacts are Exfiltrated
CCC.MLDE.TH02
Related Capabilities
| ID | Title | Description |
|---|---|---|
| CCC.MLDE.CP01 | Managed Notebook Environments | Provides fully managed notebook instances specifically designed for machine learning development, eliminating the need to manage underlying infrastructure. |
| CCC.MLDE.CP07 | Data Pipeline Integration | Supports integration with data preparation and feature engineering pipelines, including versioning of datasets and capabilities used in ML experiments. |
| CCC.MLDE.CP08 | Model Registry | Provides centralized storage and versioning for trained models, including metadata about training runs, model artifacts, and deployment history. |
Related Controls
| ID | Title | Description |
|---|---|---|
| CCC.MLDE.CN02 | Disable File Downloads on MLDE Instances | Prevent unauthorized file downloads from MLDE instances to protect sensitive data from being exfiltrated. |
| CCC.MLDE.CN07 | Restrict Public IP Access on MLDE Instances | Prevent public IP access to MLDE instances to reduce exposure to the internet and enhance security. |