Deterministic PII Sanitization in AI Training Datasets: Beyond Regex
Architecting local, distributed detection and redaction pipelines for Personally Identifiable Information (PII) to ensure GDPR compliance in massive LLM training corpora.
4 docs tagged with "gdpr"
View all tagsGDPR & The Right to be Forgotten in Neural Networks
If PII enters your model weights, standard deletion is impossible. Learn the mechanics of training data extraction and how to implement deterministic PII filtering.
Healthcare AI: Enforcing HIPAA and GDPR via Zero-Shot PII Redaction
Architectural strategies for preventing Protected Health Information (PHI) and PII leakage in healthcare RAG systems using GLiNER and hybrid ML scanning.
Preventing PII Leakage in LLM-Generated Synthetic Datasets
LLMs used for synthetic data generation frequently regurgitate real PII from their training data. Learn how to implement streaming Zero-Shot NER to sanitize datasets at scale.