Black Hat USA 2018
Black Hat, now in its 21st year, is a leading information security event that provides participants with the very latest in research, development and security trends. This year’s event includes four days of technical trainings (Aug. 4-7), followed by the two-day main conference (Aug. 8-9).
We invite you to visit the IBM Security Lounge at Booth 2104 to meet the IBM experts and see a host of solution offerings, and to attend the following conference sessions:
When: Thurs., July 9 | 1:45 p.m. – 2:45 p.m.
Where: Business Hall (Oceanside), Arsenal Theater
What: Adversarial attacks of machine learning systems have become an undisputable threat, and devising comprehensive defences against poisoning and evasion attacks by adaptive adversaries is still an open challenge. In this session we will present the Adversarial Robustness Toolbox (ART), a library which allows rapid crafting and analysis of both attacks and defence methods for machine learning models. It provides an implementation for many state-of-the-art methods for attacking and defending machine learning. Through ART, attendees will (re)discover how to attack and defend machine learning systems.
When: Thurs., July 9 | 2:30 p.m. – 3:50 p.m.
Where: Business Hall (Oceanside), Arsenal Station 9
What: Execution of an offensive payload may begin with a safe delivery of the payload to the endpoint itself. When secure connections in the enterprise are inspected, reliance only on transmission level security may not be enough to accomplish that goal. Foxtrot C2 serves one goal: safe last mile delivery of payloads and commands between the external network and the internal point of presence, traversing intercepting proxies, with the end-to-end application level encryption.
Who: Marc Ph. Stoecklin, Principal RSM and Manager, Cognitive Cybersecurity Intelligence, IBM Research
When: Thurs., July 9 | 5:00 p.m. – 6:00 p.m.
Where: South Seas ABE
What: DeepLocker is a novel class of highly targeted and evasive malware powered by artificial intelligence (AI), which is trained to reason about its environment and is able to unleash its malicious behavior only when it recognizes its target. In this session we will demonstrate how DeepLocker learns to recognize a specific target, concealing its attack payload in benign carrier applications until the intended target is identified. DeepLocker leverages various attributes for target identification, including visual, audio, geolocation and system features.