ThreatCluster

LLM Side-Channel Attack Exposes User Conversations

First seen 11 Nov 2025, 16:33 UTC •Theregister •22

Article Content

Browse articles
ThreatCluster

Microsoft researchers reported that a side-channel attack can enable unauthorized parties to infer topics discussed in large language models (LLMs). Models from providers such as Anthropic, AWS, DeepSeek, and Google remain vulnerable, affecting both personal and enterprise communications. The attack exploits indirect signals like power consumption and timing to extract sensitive information.

Ask AI about this cluster

How this analysis works

Extracted Entities

1 / 2