Name
The Often-Overlooked Privacy Problems in AI
Date & Time
Wednesday, May 22, 2024, 3:40 PM - 4:10 PM
Speakers
Description
To increase the predictive powers of large language models, it’s incumbent on businesses to train them on as much data as possible. Problems arise though when LLM outputs leak the private data on which they’ve been trained. We explore privacy-preserving methods that keep secrets secret and help organizations maintain compliance with privacy regulations such as GDPR.