AI Component

Training Data

Training data is both the model's most valuable input and its most underprotected one. Three problem classes dominate. First, poisoning: an attacker who can influence a public dataset, a web crawl, or a fine-tuning corpus can plant backdoors or biases that survive into the deployed model — BadNets-style attacks on image classifiers, trigger-phrase attacks on LLMs, and reward-hacking on RLHF datasets. Second, memorization and leakage: models can regurgitate verbatim training data, exposing PII and copyrighted content; this has driven the active New York Times v. OpenAI litigation and is a recurring GDPR concern. Third, provenance: when training data origins are unclear, downstream users inherit legal and security risk they can't assess. EU AI Act Article 10 (Data Governance) and ISO 42001 Annex A treat training-data quality as a controlled asset. Defenses: data lineage tracking, deduplication, PII scrubbing before training, and adversarial training against known trigger families.

287
Total CVEs
15
Pages
Page 2 of 15
Current
SeverityCVECVSS
CRITICALCVE-2021-359589.1
MEDIUMCVE-2021-376375.5
HIGHCVE-2021-376397.8
HIGHCVE-2021-376437.1
HIGHCVE-2021-376357.1
HIGHCVE-2021-376507.8
HIGHCVE-2021-376517.8
HIGHCVE-2021-376547.1
HIGHCVE-2021-376557.3
HIGHCVE-2021-376567.8
HIGHCVE-2021-376627.8
HIGHCVE-2021-376647.1
HIGHCVE-2021-376487.8
HIGHCVE-2021-376527.8
HIGHCVE-2021-376667.8
HIGHCVE-2021-376717.8
HIGHCVE-2021-376637.8
MEDIUMCVE-2021-376705.5
MEDIUMCVE-2021-376725.5
MEDIUMCVE-2021-376735.5

Page 2 of 15