BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//cfp.securitybsides.org.uk//bsides-london-2024//talk//MSP
 TC8
BEGIN:VTIMEZONE
TZID:GMT
BEGIN:STANDARD
DTSTART:20001029T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:GMT
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T020000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:BST
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-bsides-london-2024-MSPTC8@cfp.securitybsides.org.uk
DTSTART;TZID=GMT:20241214T112000
DTEND;TZID=GMT:20241214T113500
DESCRIPTION:The use of Generative Artificial Intelligence (AI)\, particular
 ly Large Language Models (LLMs)\, is rapidly increasing across various sec
 tors\, bringing significant advancements in automating tasks\, enhancing d
 ecision-making\, and improving user interactions. However\, this growing r
 eliance on LLMs also introduces substantial security challenges\, as these
  models are vulnerable to various cyber threats\, including adversarial at
 tacks\, data breaches\, and misinformation propagation. Ensuring the secur
 ity of LLMs is essential to maintain the integrity of their outputs\, prot
 ect sensitive information\, and build trust in AI technologies.\n\nThis ta
 lk will examine the security vulnerabilities that are inherent in Large La
 nguage Models (LLMs)\, with a particular focus on injection techniques\, c
 lient-side attacks such as Cross-Site Scripting (XSS) and HTML injection\,
  and Denial of Service (DoS) attacks. Through the simulation of these atta
 ck vectors\, the study assesses the responses of various pre-trained model
 s like GPT-3.5 Turbo and GPT-4\, revealing their susceptibility to differe
 nt forms of manipulation.\n\nThe talk will also underscore the critical ri
 sk of these vulnerabilities\, especially when exploited in a real-time cor
 porate environment\, where they can lead to significant disruptions\, unau
 thorized access\, data theft\, and compromised system integrity.
DTSTAMP:20260611T002740Z
LOCATION:Rookie track 2
SUMMARY:LLM Security: Attacks and Controls - Nazeef Khan
URL:https://cfp.securitybsides.org.uk/bsides-london-2024/talk/MSPTC8/
END:VEVENT
END:VCALENDAR