A False Sense of Privacy: Azure PII tool fails to protect 74% of MedQA data
niloofar_mire · x · 2026-09-09
Citing the paper 'A False Sense of Privacy' (arXiv:2504.21035), the author shows that PII removal and synthetic data generation don't guarantee privacy: innocuous auxiliary info like routine social activities can re-identify sensitive attributes (age, substance use history) from sanitized text. Azure's commercial PII tool fails to protect 74% of information in MedQA; differential privacy helps but hurts downstream utility. Synthetic data carries re-identification risks too.
More from Safety
- LiteLLM supply-chain attack traced to leaked master key never revoked since May — Scobleizer · 2026-10-25
- Duo Founder Shows How to Automate a YubiKey for Agentic Logins — vincent_koc · 2026-10-25
- Abliterated models let attackers run wild with no KYC while Claude won't even do basic cyber defense — jarrodwatts · 2026-10-25
- Ex-OpenAI researcher: the 'attacker diffusion lag' shielding us from AI misuse may close fast — joshua_saxe · 2026-10-25
- Covert channel math: 128-bit secret leaks in 13s, 1TB weights take centuries — vishalmisra · 2026-10-25
- Sam Altman: 'I no longer think we're gonna be able to just patch all the bugs' — GarrisonLovely · 2026-10-25