Data poisoning defense

Feeding scrapers deliberately falsified or decoy data so the datasets they harvest are corrupted and lose value for downstream model training.

grounded in: commit 6b22a13 'Plant a honeypot decoy email to poison harvester bots' — the decoy is explicitly meant to poison harvesters; data poisoning is a lasting adversarial-ML technique tied to the 'scraper/b

Connected concepts

Model collapse, Bot mitigation & scraping defense, AI training crawler, Adversarial attacks on AI