BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
DTSTART:20260308T030000
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
RRULE:FREQ=YEARLY;BYDAY=2SU;BYMONTH=3
TZNAME:PDT
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20261101T010000
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
RRULE:FREQ=YEARLY;BYDAY=1SU;BYMONTH=11
TZNAME:PST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260914T204131Z
UID:D5C72A4F-3C5A-4FF5-A8E2-2AA60F0D2A06
DTSTART;TZID=America/Los_Angeles:20260925T150000
DTEND;TZID=America/Los_Angeles:20260925T160000
DESCRIPTION:Join us for an engaging discussion on the evolving AI security 
 landscape\, focusing on large language model (LLM) security and attacks ta
 rgeting agentic AI systems. Speakers will introduce agentic AI architectur
 es\, including how AI agents use memory\, external tools\, data sources\, 
 and multi-agent communication to perform complex tasks. They will examine 
 security risks and attack types identified in frameworks such as the OWASP
  Top 10 for LLM Applications and Agentic Applications\, with particular at
 tention to direct and indirect prompt injection\, tool misuse\, data leaka
 ge\, excessive agency\, and manipulated agent behavior. The session will a
 lso explore practical defense strategies\, including input validation\, le
 ast-privilege access\, system isolation\, human oversight\, monitoring\, a
 nd secure agent design. Speakers will conclude by discussing open research
  challenges and future directions for developing secure\, reliable\, and t
 rustworthy agentic AI systems.\n\nAgenda: \nSession outline:\n\n- Agentic 
 AI Architectures\, Applications\, and Security Risks\n- OWASP Threats and 
 Attack Surfaces in LLM and Agentic AI Systems\n- Prompt Injection Attacks\
 , Tool Misuse\, and Defensive Security Controls\n- Open Research Challenge
 s and Future Directions in AI Security\n\nVirtual: https://events.vtools.i
 eee.org/m/577687
LOCATION:Virtual: https://events.vtools.ieee.org/m/577687
ORGANIZER:badruddoja@csus.edu
SEQUENCE:49
SUMMARY:Prompt Injection in Agentic AI: Attack\, Defense and the Future
URL;VALUE=URI:https://events.vtools.ieee.org/m/577687
X-ALT-DESC:Description: <br /><p>Join us for an engaging discussion on the 
 evolving AI security landscape\, focusing on large language model (LLM) se
 curity and attacks targeting agentic AI systems. Speakers will introduce a
 gentic AI architectures\, including how AI agents use memory\, external to
 ols\, data sources\, and multi-agent communication to perform complex task
 s. They will examine security risks and attack types identified in framewo
 rks such as the OWASP Top 10 for LLM Applications and Agentic Applications
 \, with particular attention to direct and indirect prompt injection\, too
 l misuse\, data leakage\, excessive agency\, and manipulated agent behavio
 r. The session will also explore practical defense strategies\, including 
 input validation\, least-privilege access\, system isolation\, human overs
 ight\, monitoring\, and secure agent design. Speakers will conclude by dis
 cussing open research challenges and future directions for developing secu
 re\, reliable\, and trustworthy agentic AI systems.</p><br /><br />Agenda:
  <br /><p dir="auto">Session outline:</p>\n<ul>\n<li>Agentic AI Architectu
 res\, Applications\, and Security Risks</li>\n<li>OWASP Threats and Attack
  Surfaces in LLM and Agentic AI Systems</li>\n<li>Prompt Injection Attacks
 \, Tool Misuse\, and Defensive Security Controls</li>\n<li>Open Research C
 hallenges and Future Directions in AI Security</li>\n</ul>\n<p>&nbsp\;</p>
 \n<div>\n<p>&nbsp\;</p>\n<p>&nbsp\;</p>\n</div>
END:VEVENT
END:VCALENDAR