BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Towards the Offensive and Defensive Use of AI
X-WR-TIMEZONE:Central Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260908T060240Z
UID:tag:localist.com\,2008:EventInstance_52507686146725
DTSTART:20260417T160000Z
DTEND:20260417T170000Z
DESCRIPTION:Gang Wang\n\nThe fast development of artificial intelligence (A
 I) presents new opportunities for both offensive and defensive use of AI i
 n the security domain. In this talk\, I will present several case studies 
 to demonstrate (1) the emerging security threats introduced by AI-powered 
 techniques\, and (2) the challenges to applying AI for security defense. F
 irst\, I will discuss new attacks enabled by AI models\, using an example 
 of face de-anonymization attacks. I will show how specially customized Dif
 fusion models can be used to restore and re-identify Gaussian-blurred face
  images\, compromising the intended privacy protection. Second\, I will di
 scuss the challenges and possible solutions to secure AI-assisted coding w
 hile defending against vulnerable/malicious code generation. I will descri
 be our recent work to align Code LLMs to perform safety reasoning via rule
  learning to defend against the aforementioned threats. I will also share 
 our experience of developing and testing this model as part of the Amazon 
 Nova AI Challenge (2025). Finally\, I will discuss general and open challe
 nges in offensive and defensive use of AI for security.
LOCATION:Jubel 121
SUMMARY:Towards the Offensive and Defensive Use of AI
URL;VALUE=URI:https://happenings.washu.edu/event/towards-the-offensive-and-
 defensive-use-of-ai
CATEGORIES:Seminar/Colloquia
END:VEVENT
END:VCALENDAR