Can automated processing make privacy notice/choice more effective ...
summary
Project builds annotation tools, uses crowdsourcing + machine learning and NLP to extract concrete data practices from privacy policies, trains classifiers on human-labeled paragraphs and highlights likely relevant text. Prototype identifies collection of contact information reliably, flags third‑party sharing harder cases for human review, and already powers a portal covering 6,000 financial notices. Results scale toward hundreds–thousands of sites but hit limits: ambiguity, scattered clauses, notice‑and‑choice legal gaps, and need for backend audits to verify actual data handling.
AI-written summary · about
chapters (4)
- 0:00Start
- 13:40Cameron Russell Fordham University (US)
- 20:25Aaron Burstein Federal Trade Commission (US)
- 35:22Alexander Dix Berlin Data Protection Commissioner (DE)
related talks
25mPolisis: Automated Analysis and Presentation of Privacy Policies Using Deep Learning
15mTurning Privacy Risk Assessment Into 20 Questions for Developers
16mPrivacy Review for Non-Maniacs
28mPrivacy Engineering Networking Break AMA
29mDeveloping Responsive Privacy Programs and Driving Impact
30mAddressing the Privacy Paradox – Setting Consent & Preference Management Into Practice
16mAutomating Technical Privacy Reviews Using LLMs
01h 17mPrivacy and New Technologies