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 12 of 15
Current
SeverityCVECVSS
MEDIUMCVE-2026-691466.5
HIGHGHSA-wg9g-w2j2-8pgr7.8
HIGHGHSA-rghg-q7wp-9767-
HIGHGHSA-qxq5-qhx6-94qw7.8
LOWCVE-2026-494313.3
HIGHGHSA-p77j-g7h5-r2vw-
HIGHCVE-2026-763367.1
HIGHCVE-2026-762547.5
MEDIUMCVE-2026-763415.4
UNKNOWNCVE-2026-18286-
HIGHCVE-2026-760738.8
LOWCVE-2026-98052.7
UNKNOWNCVE-2026-69129-
MEDIUMCVE-2026-822905.3
MEDIUMCVE-2026-76804.3
HIGHCVE-2026-851798.5
HIGHCVE-2026-852117.7
UNKNOWNCVE-2026-85063-
MEDIUMCVE-2026-856706.5
MEDIUMCVE-2026-706266.2

Page 12 of 15