{"id":147923,"date":"2025-11-10T23:11:54","date_gmt":"2025-11-10T21:11:56","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/ataka-whisper-leak-po-opredeleniyu-tematiki-obshheniya-s-ai-chatbotami-cherez-analiz-tls-trafika"},"modified":"2025-11-10T23:11:54","modified_gmt":"2025-11-10T21:11:56","slug":"ataka-whisper-leak-po-opredeleniyu-tematiki-obshheniya-s-ai-chatbotami-cherez-analiz-tls-trafika","status":"publish","type":"post","link":"https:\/\/prohoster.info\/hu\/blog\/news\/ataka-whisper-leak-po-opredeleniyu-tematiki-obshheniya-s-ai-chatbotami-cherez-analiz-tls-trafika","title":{"rendered":"A Whisper Leak t\u00e1mad\u00e1s az AI chatbotokkal folytatott kommunik\u00e1ci\u00f3 t\u00e9m\u00e1j\u00e1nak meghat\u00e1roz\u00e1s\u00e1ra TLS forgalom elemz\u00e9s\u00e9vel","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
A Microsoft security researcher has developed a Whisper Leak side-channel attack technique that allows the classification of topics in requests to large language model-based services through passive analysis of encrypted traffic transmitted via a TLS connection. Information about the size of network packets and delays between their transmission is sufficient to determine request topics for AI chatbots with over 98% accuracy. In practice, the proposed method can be used to identify certain request topics in user transit traffic, such as attempts to obtain information about illegal activities, without decrypting the content. <\/p>\n
Adatkinyer\u0151 eszk\u00f6zk\u00e9szlet el\u00e9rhet\u0151 a forgalmi ment\u00e9sekb\u0151l val\u00f3 adatok kinyer\u00e9s\u00e9re, a modell oktat\u00e1s\u00e1ra \u00e9s a m\u00f3dszer tesztel\u00e9s\u00e9re a GitHub-on. A t\u00e1mad\u00e1s lehet\u0151s\u00e9g\u00e9t 28 n\u00e9pszer\u0171, nagy nyelvi modell sz\u00e1m\u00e1ra demonstr\u00e1lt\u00e1k, a legnagyobb gy\u00e1rt\u00f3k k\u00f6z\u00fcl. P\u00e9ld\u00e1ul a \"p\u00e9nzmos\u00e1s\" t\u00e9m\u00e1j\u00fa k\u00e9rd\u00e9sek azonos\u00edt\u00e1s\u00e1nak pontoss\u00e1ga sok AI-szolg\u00e1ltat\u00e1s eset\u00e9ben 100% volt, amikor az elemzett forgalomban 1 keresett k\u00e9rd\u00e9s \u00e9s 10 000, a sz\u00fcks\u00e9ges t\u00e9m\u00e1hoz nem kapcsol\u00f3d\u00f3 k\u00e9rd\u00e9s volt jelen.
<\/center> <\/p>\n
The cause of the information leak is that models generate a response to a request step by step, one token at a time, with each step using the previous token as context to determine the next most likely word or phrase. Accordingly, a separate network packet is sent for each token, and the delay between packets corresponds to the delay in the model's determination of the next token. <\/p>\n
In TLS, if data compression is not used, the ciphertext size equals the size of the plaintext plus a constant. When creating a model that correlates the desired sets of tokens with packet sizes and delays between their transmission, it is possible to fairly accurately determine the presence of desired topics in the traffic. During the study, three versions of such machine learning models were prepared, based on the LightGBM, Bi-LSTM, and BERT neural network architectures. Experiments were conducted for each model to determine the desired topic by analyzing only the packet sizes, only the delays between packets, and both criteria.
<\/center> <\/p>\n
Az AI-szolg\u00e1ltat\u00e1sok fejleszt\u0151i sz\u00e1m\u00e1ra javasolt, hogy a keres\u00e9si t\u00e9ma passz\u00edv elemz\u00e9s\u00e9nek hat\u00e9konys\u00e1g\u00e1nak cs\u00f6kkent\u00e9se \u00e9rdek\u00e9ben v\u00e9letlenszer\u0171 kieg\u00e9sz\u00edt\u0151 kit\u00f6lt\u00e9st alkalmazzanak, bufferezz\u00e9k a tokenek \u00e1tvitel\u00e9t, vagy hamis csomagok behelyettes\u00edt\u00e9s\u00e9vel dolgozzanak. \n Forr\u00e1s: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"
![\"A](\"\/wp-content\/uploads\/2025\/11\/9e2db8a564ae07b2ae7ced2d2f744e46.png\")
<\/center> <\/p>\n![\"A](\"\/wp-content\/uploads\/2025\/11\/d6be536d5b8814af45a42317c9454a44.png\")
<\/center> <\/p>\n