Beyond Memorization: Violating Privacy Via Inference with Large Language Models

Understand

Current privacy research on large language models (LLMs) primarily focuses on the issue of extracting memorized training data.

• At the same time, models' inference capabilities have increased drastically.
• This raises the key question of whether current LLMs could violate individuals' privacy by inferring personal attributes from text given at inference time.
• In this work, we present the first comprehensive study on the capabilities of pretrained LLMs to infer personal attributes from text.