Combining Privacy and AI
summary
Brighter AI presents a deep‑learning anonymization that replaces faces and license plates in camera feeds with realistic synthetic objects so companies can train models on usable images while meeting GDPR and CCPA constraints. Cameras remain usable for automotive, retail and smart‑city analytics; systems that normally extract up to 20,000 facial keypoints now see synthetic substitutes with only 16% re‑identification rate against a commercial matcher. Pipeline combines synthetic replacement, data augmentation and optional federated workflows to preserve utility.
AI-written summary · about
chapters (14)
- 0:00Intro
- 1:50Cameras
- 2:32Facial Recognition
- 3:13Facial Recognition Abuse
- 4:09Data Privacy Regulation
- 6:22AI Privacy Dilemma
- 7:07Privacy Concepts
- 8:09Decentralization
- 9:19Privacy by default
- 10:05Privacy by design
- 11:19Synthetic data
- 12:19Data augmentation
- 13:19Privacy dilemma
- 14:53Synthetic objects
related talks
19mGDPR, AI & use of Video data
20mWIP: Blurred Lines -- A GDPR-Compliant Framework for Anonymising Automotive...
40mNew Face, Who Dis? Protecting Privacy in a World of Surveillance
20mSynthetic Data for Privacy, AI Fairness and Democratizing Data Access
01h 15mHOW TO RECONCILE FACIAL RECOGNITION TECHNOLOGIES WITH CONSUMERS' PRIVACY
17mDon't Stop at Data Democratization!
28mInsights, Not Secrets: How Synthetic Data Empowers Privacy
54mOpen Data vs. Privacy - How Synthetic Data Can Democratize AI Innovation