Threatline

LLM side-channel attack could allow snoops to guess what you're talking about

Encryption protects content, not context Mischief-makers can guess the subjects being discussed with LLMs using a side-channel attack, according to Microsoft researchers. They told that models from some providers, including Anthropic, AWS, DeepSeek, and Google, haven't been fixed, putting both personal users and enterprise communications at risk.… The Register