Banana Navy
Catalog FR
Lab · Threat modeling IA · Fiche détaillée

Data and supply chain poisoning

untrusted content durably contaminates the knowledge base (RAG), the model (poisoned model) or the supply chain (vulnerable dependency, malicious package, outdated detector), influencing every future answer.

SheetF19
CategoryAI5-TB7 (data poisoning / RAG poisoning / supply chain)
Layers20 · Model & Supply Chain, 14 · LLM Interpretation
Systemvoicebot IA

only approved ingestion writes to the base (no caller speech re-injected as-is), SBOM, digests and signatures of model artifacts, CVE scanning, adversarial benchmarks.

The threat

untrusted content durably contaminates the knowledge base (RAG), the model (poisoned model) or the supply chain (vulnerable dependency, malicious package, outdated detector), influencing every future answer.

Angle mortWhy classic frameworks miss it

the database here is knowledge that expresses itself in the answers; poisoning is slow, diffuse, and looks like neither SQL injection nor a network compromise.

MitigationProposed approach

only approved ingestion writes to the base (no caller speech re-injected as-is), SBOM, digests and signatures of model artifacts, CVE scanning, adversarial benchmarks.

The proposed control
no unapproved content enters the base; model artifacts verified by digest.

Expected evidence
trace that no unapproved content can enter the KB and that model artifacts are digest-verified.

SourcesReferences and public research

OWASP GenAIOWASP GenAI LLM04:2025 Data and Model Poisoning · OWASP GenAI LLM03:2025 Supply Chain
Recherche publiquePublic research sources: MITRE ATLAS 2026.07 (verified technique mapping), OWASP GenAI (OWASP), and the public risk-voicebot (aivansoul/risk-voicebot) template defining the 20 checkpoints. No client registry data: generic sheet, no rating, no verdict.
couche 20 · Model & Supply Chaincouche 14 · LLM Interpretation

Explore the 20 security layers

MITRE ATLAS 2026.07 · OWASP GenAI · risk-voicebot