Selling training data without anonymization risks re-identification, founder warns
annetgriffin · x · 2026-10-02
The author warns businesses packaging training data for sale: un-anonymized data can be retroactively identified, mapping users to accounts, businesses, or private dealings, and AI tools can easily build "digital twins" from it. Fingerprinting is improving fast so sellers can't hide from accountability; also, winding-down companies can't sell data unless permitted by their TOS and liquidation assets.
More from Safety
- The standing argument in LASST v. OpenAI, the Hugging Face case — hoofnagle · 2026-10-02
- Swarms adds 0-100 Security Scores to every marketplace prompt via SkillScanner — KyeGomezB · 2026-10-02
- Buck's LessWrong essay on tiers of safety buy-in inside AI labs — anpaure · 2026-10-02
- AI Explained: OpenAI warns controlling frontier models is now 'hell' as GPT-6.1 Astra gets shelved — AI Explained · 2026-10-02
- David Krueger's post-AGI workshop talk rejects 'managing the transition' framing — DavidSKrueger · 2026-10-02
- Federal judge blocks New York's ban on algorithmic rent pricing — news-10 · 2026-10-02