Skip to main content

CCC.GenAI.TH02: Data Poisoning

Threat ID:CCC.GenAI.TH02
Title:Data Poisoning
Description:

Data poisoning occurs when training, fine-tuning or embedding data is tampered with in order to modify the model's behaviour, for example steering it towards specific outputs, degrading performance or introducing backdoors.

Related Capabilities

IDTitleDescription
CCC.Core.F02Encryption at Rest Enabled by DefaultThe service automatically encrypts all data using industry-standard cryptographic protocols prior to being written to a storage medium.
CCC.Core.F06Access ControlThe service automatically enforces user configurations to restrict or allow access to a specific component or a child resource based on factors such as user identities, roles, groups, or attributes.
CCC.GenAI.F03Embedding Model SelectionAbility to select a foundation model used for tasks like semantic search, clustering, and document similarity by converting text into vector embeddings.
CCC.GenAI.F06Customizable Model SelectionProvide users the ability to fine-tune models with their own data.
CCC.GenAI.F21Generate ContentAbility to generate a response given a foundation model, parameter values, and a prompt.
CCC.GenAI.F22Data ControlEnsures prompts, model outputs, embeddings, and training data fed by customers are not used to train foundation models.
CCC.GenAI.F24Content ModerationEnsure the service detects and filters abusive, harmful, and sensitive information to ensure responsible and safe use of the service.

External Mappings

Reference IDEntry IDStrengthRemarks
FINOS-AIGF
AIR-SEC-009
0
Data Poisoning
SAIF
DP
0
Data Poisoning
OWASP-LLM-TOP10
LLM04:2025
0
Data and Model Poisoning
MITRE-ATLAS
AML.T0020
0
Poison Training Data
MITRE-ATLAS
AML.T0070
0
RAG Poisoning

Controls

IDTitleObjectiveControl FamilyThreat MappingsGuideline MappingsAssessment Requirements
CCC.GenAI.C03Data Provenance and Source VettingEnsure that all data for training, fine-tuning or RAG comes from trusted, approved sources and is authorised for the intended purposes in order to prevent the initial introduction of malicious content or leaked sensitive data. Data
2
3
2
CCC.GenAI.C04Sanitisation of Ingested DataValidate and sanitise all data ingested by GenAI systems from extenal sources or internal knowledge bases, whether for training, conversion to vector embeddings, or real-time retireval, in order to remove or redact poisoned or sensitive data before further processing. Data
2
3
2
CCC.GenAI.C08Quality Control and Red TeamingEstablish a formal program for quality evaluation and adversarial testing (red teaming) to ensure GenAI system meet all business, quality, security and compliance requirements before getting deployed into production environments. Model Assurance and Evaluation
5
5
2