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 13 of 15
Current
SeverityCVECVSS
HIGHCVE-2026-786827.5
MEDIUMCVE-2026-819164.3
HIGHCVE-2026-907778.8
HIGHCVE-2026-917718.8
UNKNOWNCVE-2026-62997-
HIGHCVE-2026-968048.8
HIGHCVE-2026-1008417.8
HIGHCVE-2026-1008437.8
HIGHCVE-2026-1008447.8
HIGHCVE-2026-1008457.8
HIGHCVE-2026-1008468.8
HIGHCVE-2026-933488.1
HIGHCVE-2026-73975-
CRITICALCVE-2026-1040199.0
HIGHCVE-2026-1048518.8
HIGHCVE-2026-1093607.5
HIGHCVE-2026-1093617.3
MEDIUMCVE-2026-1102485.3
HIGHCVE-2026-1041038.1
MEDIUMCVE-2026-1111706.5

Page 13 of 15