LLMs & Online Privacy: Deanonymization Advances
Recent advances in large language models (LLMs) have enabled the rapid and accurate re-identification of users who previously operated under pseudonyms online. These models can process vast amounts of publicly available textual information to link anonymous profiles with real identities, often achieving success rates comparable to expert investigators. A collaborative effort involving ETH...
0 Comentários 0 Compartilhamentos 4 Visualizações 0 Anterior