Automatisation de l'analyse de binaires : de la collecte source ouverte à la Threat Intel
De nombreuses sources ouvertes de binaires, et particulièrement de malware ont émergé dans le paysage ces dernières années. Et leur qualité n’a rien à envier aux sources commerciales comme le soulignait Thibaut Binetruy (HMiser, CERT Société Generale, 2020), “Integrating operational threat intel in your defense mechanisms doesn’t mean buying Threat Intel. You can start by using the [mass] of open source indicators available for free.”. Certaines sont mises à disposition par des sources officielles (Abuse.ch, alimenté entre autre par le CERT national Suisse), d’autres beaucoup plus obscures voire anonymes (VirusShare, Vx-underground…). Le panorama que nous en avons dressé souligne la grande disparité qualitative et quantitative de ces sources. Il nous a fallu prendre en compte cette diversité dans le cadre de nos travaux de recherche, afin de rendre possible l’analyse quotidienne des corrélations inter- et intra-familles de malwares à grande échelle. Ces travaux permettent une application sur des cas concrets tels que Babuk, Ryuk et Conti. Nous avons ainsi pu mettre en évidence les liens sur les échantillons de ces familles grâce à l’identification immédiate de corrélations, complétée par une analyse manuelle qui a ainsi permis de confirmer précisément la généalogie des échantillons.
Many open feeds of binary files, especially malware, have emerged in the landscape in recent years. And their quality has nothing to envy to commercial sources as emphasized by Thibaut Binetruy (HMiser, CERT Société Generale, 2020), “Integrating operational threat intel in your defense mechanisms doesn’t mean buying Threat Intel. You can start by using the [mass] of open source indicators available for free”. Some are made available by official sources (Abuse.ch, supplied among others by the Swiss national CERT), others much more obscure or even anonymous (VirusShare, Vx-underground…). The panorama that we have drawn up underlines the great qualitative and quantitative disparity of these sources. We had to take this diversity into account in the context of our research, in order to make possible the daily analysis of inter- and intra-family correlations of malware at large scale. These works allow an application on concrete cases such as Babuk, Ryuk and Conti. We were able to highlight the links between the samples in those families thanks to the immediate identification of correlations, supplemented by a manual analysis which thus made it possible to precisely confirm the genealogy of the samples.
Tue 16 NovDisplayed time zone: Brussels, Copenhagen, Madrid, Paris change
13:30 - 15:00 | Cyber Threat IntelligenceCall for Papers at Grand Auditorium Chair(s): Gurvan LE GUERNIC DGA MI & Université de Rennes 1 | ||
13:30 30mTalk | La Threat Intelligence comme vecteur d’automatisation de la Cyberdéfense Call for Papers Media Attached File Attached | ||
14:00 20mTalk | Automatisation de l'analyse de binaires : de la collecte source ouverte à la Threat Intel Call for Papers Media Attached | ||
14:20 20mTalk | Automated Risk Analysis of a Vulnerability Disclosure Using Active Learning Call for Papers Media Attached | ||
14:40 20mTalk | Attack Forecast and Prediction Call for Papers Florian Kaiser Karlsruhe Institute of Technology, Tobias Budig Karlsruhe Institute of Technology, Elisabeth Goebel Karlsruhe Institute of Technology, Tessa Fischer Karlsruhe Institute of Technology, Jurek Muff Karlsruhe Institute of Technology, Marcus Wiens Karlsruhe Institute of Technology, Frank Schultmann Karlsruhe Institute of Technology Media Attached |