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:20260510T190702Z
UID:98409658-D1A0-45F9-960A-6A81057122AA
DTSTART;TZID=America/Los_Angeles:20260526T164500
DTEND;TZID=America/Los_Angeles:20260526T190000
DESCRIPTION:We are excited to continue the Orange County Computer Society (
 OCCS) Global Emerging Technologies and Artificial Intelligence (GET-AI) Se
 ries—a monthly platform focused on transformative innovations in compute
 r science and technology. Hosted by the IEEE Orange County Computer Societ
 y Chapter\, this series brings together professionals\, students\, and tec
 h enthusiasts to explore the cutting edge of what’s possible.\n\nFollowi
 ng a highly engaging April session on Generative AI\, where we explored LL
 Ms\, RAG\, Agents\, MCP\, and hands-on AI application development\, we are
  excited to bring you our May Tech Talk on “Security in AI.”\n--------
 -------------------------------------------------------\n\n🔒 May Focus:
  Securing Generative AI\n\nAs AI systems evolve—from traditional models 
 to LLM-powered agents interacting with enterprise systems and real-world t
 ools—they introduce powerful capabilities along with new security challe
 nges\, including:\n\n- Data leakage and prompt injection\n- Model misuse a
 nd unauthorized access\n- Risks in agent-driven automation\n- Governance a
 nd compliance concerns\n\nThis session combines technical insights and pra
 ctical demonstrations to explore how to build secure\, trustworthy AI syst
 ems at scale.\n-----------------------------------------------------------
 ----\n\nSession 1: Intelligent Attack Detection &amp; Provenance (45 mins)\n\n
 Modern enterprises generate massive\, fragmented logs\, making it difficul
 t to derive meaningful security insights.\n\nThis session explores how AI 
 enhances detection and forensic analysis:\n\n- Graph-Based Intrusion Detec
 tion\nUse unsupervised graph learning to uncover multi-step attacks in net
 work activity\n- LLM-Powered Security Intelligence\nConvert low-level aler
 ts into high-level\, actionable insights for faster response\n\n👉 Takea
 way: Move from fragmented alerts to intelligent\, end-to-end attack unders
 tanding\n---------------------------------------------------------------\n
 \nSession 2: Securing AI Agents — MCP Threats &amp; Defense (45 mins)\n\nAs 
 AI agents integrate with tools\, APIs\, and external systems\, they introd
 uce new attack surfaces.\n\nThis session includes a live demo of how agent
 s can be compromised—and secured:\n\n- Understanding MCP Architectures\n
 How agents invoke tools and why trust boundaries blur\n- Live Demo: Tool P
 oisoning &amp; Agent Manipulation\nSee how adversarial inputs can:\n- Manipula
 te agent behavior\n- Trigger unintended actions\n- Lead to data exfiltrati
 on\n\n- Layered Security Framework\nPractical defenses:\n- Tool authentica
 tion\n- Response sanitization\n- Schema validation\n- Context isolation\n\
 n- Real-Time Evaluation\nPrevent attacks without impacting performance\n\n
 👉 Takeaway: Practical strategies to secure AI agents in enterprise envi
 ronments\n---------------------------------------------------------------\
 n\nAbout the Organizer\n\nPradyumna Kodgi\nPrincipal Product Manager | Ora
 cle Health &amp; AI\nIEEE Senior Member | Vice Chair\, IEEE EMBS – Orange Co
 unty\nMember\, IEEE AI Agentic Systems &amp; AI Policy Committees\n\n📍 Cali
 fornia\, USA\n📧 pkodgi@ieee.org\n🔗 linkedin.com/in/pkodgi\n\nCo-spon
 sored by: Pradyumna Kodgi\n\nSpeaker(s): Zhou\, Sreekanth\n\nAgenda: \nSec
 uring AI: From Innovation to Resilience\n\nAI is rapidly transforming how 
 we build intelligent systems—but as capabilities grow\, so do security r
 isks. From LLM-powered agents to tool-integrated architectures\, the quest
 ion is no longer just what AI can do—but how do we secure it?\n\nIn this
  interactive session\, we cut through the noise and break down AI security
  in practical\, real-world terms—so you can understand not just the risk
 s\, but how to defend against them.\n-------------------------------------
 --------------------------\n\n🔍 What You’ll Explore\n\n- How modern A
 I systems (LLMs\, agents\, MCP) introduce new attack surfaces\n- The shift
  from traditional security to AI-driven threat models\n- Key security conc
 epts—explained clearly and practically\n- Real-world attack scenarios an
 d emerging threat patterns\n----------------------------------------------
 -----------------\n\n💡 What Makes This Session Different\n\nThis isn’
 t just theory—you’ll see AI systems under attack and defense in action
 .\n\nThrough a live\, end-to-end demonstration\, we’ll show how AI agent
 s can be manipulated—and how layered security approaches can prevent the
 se attacks in real time.\n------------------------------------------------
 ---------------\n\n🛠️ Practical Takeaways\n\nYou’ll walk away with 
 actionable strategies and frameworks you can apply immediately\, including
 :\n\n- Securing AI agents interacting with external tools\n- Validating an
 d sanitizing untrusted inputs\n- Designing trust boundaries in AI-driven a
 rchitectures\n------------------------------------------------------------
 ---\n\n🎯 Who Should Attend\n\n- Security professionals and architects w
 orking with AI systems\n- Engineers and developers building AI/LLM-based a
 pplications\n- Product managers and leaders driving AI adoption\n- Anyone 
 interested in understanding AI risks and defenses\n-----------------------
 ----------------------------------------\n\n✨ What You’ll Walk Away Wi
 th\n\n- A clear understanding of emerging AI security risks\n- Practical k
 nowledge of how to secure AI agents and systems\n- Real-world insights int
 o attack prevention and defense strategies\n------------------------------
 ---------------------------------\n\nAs AI systems become more autonomous 
 and integrated into enterprise workflows\, security becomes foundational
 —not optional. This session will equip you with the mindset and tools to
  build AI systems you can trust.\n\nVirtual: https://events.vtools.ieee.or
 g/m/557806
LOCATION:Virtual: https://events.vtools.ieee.org/m/557806
ORGANIZER:pkodgi@ieee.org
SEQUENCE:69
SUMMARY:2026 GET-AI SERIES: 2 . Trust in AI Systems: Detecting\, Defending\
 , and Securing Intelligent Agents
URL;VALUE=URI:https://events.vtools.ieee.org/m/557806
X-ALT-DESC:Description: &lt;br /&gt;&lt;p data-start=&quot;111&quot; data-end=&quot;530&quot;&gt;We are exc
 ited to continue the &lt;strong data-start=&quot;142&quot; data-end=&quot;256&quot;&gt;Orange County
  Computer Society (OCCS) Global Emerging Technologies and Artificial Intel
 ligence (GET-AI) Series&lt;/strong&gt;&amp;mdash\;a monthly platform focused on tran
 sformative innovations in computer science and technology. Hosted by the I
 EEE Orange County Computer Society Chapter\, this series brings together p
 rofessionals\, students\, and tech enthusiasts to explore the cutting edge
  of what&amp;rsquo\;s possible.&lt;/p&gt;\n&lt;p data-start=&quot;532&quot; data-end=&quot;749&quot;&gt;Follow
 ing a highly engaging &lt;strong data-start=&quot;560&quot; data-end=&quot;594&quot;&gt;April sessio
 n on Generative AI&lt;/strong&gt;\, where we explored LLMs\, RAG\, Agents\, MCP\
 , and hands-on AI application development\, we are excited to bring you ou
 r &lt;strong data-start=&quot;711&quot; data-end=&quot;749&quot;&gt;May Tech Talk on &amp;ldquo\;Securit
 y in AI.&amp;rdquo\;&lt;/strong&gt;&lt;/p&gt;\n&lt;hr data-start=&quot;751&quot; data-end=&quot;754&quot;&gt;\n&lt;h2 d
 ata-section-id=&quot;7mtamf&quot; data-start=&quot;756&quot; data-end=&quot;795&quot;&gt;🔒 May Focus: Se
 curing Generative AI&lt;/h2&gt;\n&lt;h2 data-section-id=&quot;1f9jw5j&quot; data-start=&quot;894&quot; 
 data-end=&quot;937&quot;&gt;&lt;img src=&quot;https://events.vtools.ieee.org/vtools_ui/media/di
 splay/b7fe099e-6f6f-40dc-b049-84356c923e00&quot; width=&quot;1032&quot; height=&quot;534&quot;&gt;&lt;/h2
 &gt;\n&lt;p data-start=&quot;797&quot; data-end=&quot;1008&quot;&gt;As AI systems evolve&amp;mdash\;from tr
 aditional models to &lt;strong data-start=&quot;845&quot; data-end=&quot;924&quot;&gt;LLM-powered ag
 ents interacting with enterprise systems and real-world tools&lt;/strong&gt;&amp;mda
 sh\;they introduce powerful capabilities along with new security challenge
 s\, including:&lt;/p&gt;\n&lt;ul data-start=&quot;1010&quot; data-end=&quot;1164&quot;&gt;\n&lt;li data-secti
 on-id=&quot;yn439c&quot; data-start=&quot;1010&quot; data-end=&quot;1047&quot;&gt;Data leakage and prompt i
 njection&lt;/li&gt;\n&lt;li data-section-id=&quot;15gtg3k&quot; data-start=&quot;1048&quot; data-end=&quot;1
 088&quot;&gt;Model misuse and unauthorized access&lt;/li&gt;\n&lt;li data-section-id=&quot;1vq2f
 yu&quot; data-start=&quot;1089&quot; data-end=&quot;1125&quot;&gt;Risks in agent-driven automation&lt;/li
 &gt;\n&lt;li data-section-id=&quot;f3fezf&quot; data-start=&quot;1126&quot; data-end=&quot;1164&quot;&gt;Governan
 ce and compliance concerns&lt;/li&gt;\n&lt;/ul&gt;\n&lt;p data-start=&quot;1166&quot; data-end=&quot;130
 8&quot;&gt;This session combines &lt;strong data-start=&quot;1188&quot; data-end=&quot;1239&quot;&gt;technic
 al insights and practical demonstrations&lt;/strong&gt; to explore how to build 
 &lt;strong data-start=&quot;1264&quot; data-end=&quot;1307&quot;&gt;secure\, trustworthy AI systems 
 at scale&lt;/strong&gt;.&lt;/p&gt;\n&lt;hr data-start=&quot;1310&quot; data-end=&quot;1313&quot;&gt;\n&lt;h2 data-s
 ection-id=&quot;nndyfq&quot; data-start=&quot;1315&quot; data-end=&quot;1380&quot;&gt;Session 1: Intelligen
 t Attack Detection &amp;amp\; Provenance (45 mins)&lt;/h2&gt;\n&lt;p data-start=&quot;1382&quot; 
 data-end=&quot;1495&quot;&gt;Modern enterprises generate massive\, fragmented logs\, ma
 king it difficult to derive meaningful security insights.&lt;/p&gt;\n&lt;p data-sta
 rt=&quot;1497&quot; data-end=&quot;1567&quot;&gt;This session explores how AI enhances detection 
 and forensic analysis:&lt;/p&gt;\n&lt;ul data-start=&quot;1569&quot; data-end=&quot;1823&quot;&gt;\n&lt;li da
 ta-section-id=&quot;1n1fup6&quot; data-start=&quot;1569&quot; data-end=&quot;1694&quot;&gt;&lt;strong data-sta
 rt=&quot;1571&quot; data-end=&quot;1606&quot;&gt;Graph-Based Intrusion Detection&lt;/strong&gt;&lt;br data
 -start=&quot;1606&quot; data-end=&quot;1609&quot;&gt;Use unsupervised graph learning to uncover m
 ulti-step attacks in network activity&lt;/li&gt;\n&lt;li data-section-id=&quot;1xvsq5t&quot; 
 data-start=&quot;1696&quot; data-end=&quot;1823&quot;&gt;&lt;strong data-start=&quot;1698&quot; data-end=&quot;1735
 &quot;&gt;LLM-Powered Security Intelligence&lt;/strong&gt;&lt;br data-start=&quot;1735&quot; data-end
 =&quot;1738&quot;&gt;Convert low-level alerts into high-level\, actionable insights for
  faster response&lt;/li&gt;\n&lt;/ul&gt;\n&lt;p data-start=&quot;1825&quot; data-end=&quot;1915&quot;&gt;👉 &lt;e
 m data-start=&quot;1828&quot; data-end=&quot;1915&quot;&gt;Takeaway: Move from fragmented alerts 
 to intelligent\, end-to-end attack understanding&lt;/em&gt;&lt;/p&gt;\n&lt;hr data-start=
 &quot;1917&quot; data-end=&quot;1920&quot;&gt;\n&lt;h2 data-section-id=&quot;34kwv5&quot; data-start=&quot;1922&quot; da
 ta-end=&quot;1988&quot;&gt;Session 2: Securing AI Agents &amp;mdash\; MCP Threats &amp;amp\; De
 fense (45 mins)&lt;/h2&gt;\n&lt;p data-start=&quot;1990&quot; data-end=&quot;2088&quot;&gt;As AI agents in
 tegrate with tools\, APIs\, and external systems\, they introduce new atta
 ck surfaces.&lt;/p&gt;\n&lt;p data-start=&quot;2090&quot; data-end=&quot;2173&quot;&gt;This session includ
 es a &lt;strong data-start=&quot;2114&quot; data-end=&quot;2127&quot;&gt;live demo&lt;/strong&gt; of how a
 gents can be compromised&amp;mdash\;and secured:&lt;/p&gt;\n&lt;ul data-start=&quot;2175&quot; da
 ta-end=&quot;2699&quot;&gt;\n&lt;li data-section-id=&quot;1whldox&quot; data-start=&quot;2175&quot; data-end=&quot;
 2272&quot;&gt;&lt;strong data-start=&quot;2177&quot; data-end=&quot;2212&quot;&gt;Understanding MCP Architec
 tures&lt;/strong&gt;&lt;br data-start=&quot;2212&quot; data-end=&quot;2215&quot;&gt;How agents invoke tool
 s and why trust boundaries blur&lt;/li&gt;\n&lt;li data-section-id=&quot;1i1z3ow&quot; data-s
 tart=&quot;2274&quot; data-end=&quot;2459&quot;&gt;&lt;strong data-start=&quot;2276&quot; data-end=&quot;2326&quot;&gt;Live
  Demo: Tool Poisoning &amp;amp\; Agent Manipulation&lt;/strong&gt;&lt;br data-start=&quot;23
 26&quot; data-end=&quot;2329&quot;&gt;See how adversarial inputs can:\n&lt;ul data-start=&quot;2365&quot;
  data-end=&quot;2459&quot;&gt;\n&lt;li data-section-id=&quot;2vdai7&quot; data-start=&quot;2365&quot; data-end
 =&quot;2394&quot;&gt;Manipulate agent behavior&lt;/li&gt;\n&lt;li data-section-id=&quot;i1y43f&quot; data-
 start=&quot;2397&quot; data-end=&quot;2427&quot;&gt;Trigger unintended actions&lt;/li&gt;\n&lt;li data-sec
 tion-id=&quot;15ll3ii&quot; data-start=&quot;2430&quot; data-end=&quot;2459&quot;&gt;Lead to data exfiltrat
 ion&lt;/li&gt;\n&lt;/ul&gt;\n&lt;/li&gt;\n&lt;li data-section-id=&quot;4shnfj&quot; data-start=&quot;2461&quot; dat
 a-end=&quot;2619&quot;&gt;&lt;strong data-start=&quot;2463&quot; data-end=&quot;2493&quot;&gt;Layered Security Fr
 amework&lt;/strong&gt;&lt;br data-start=&quot;2493&quot; data-end=&quot;2496&quot;&gt;Practical defenses:\
 n&lt;ul data-start=&quot;2520&quot; data-end=&quot;2619&quot;&gt;\n&lt;li data-section-id=&quot;16gvdow&quot; dat
 a-start=&quot;2520&quot; data-end=&quot;2543&quot;&gt;Tool authentication&lt;/li&gt;\n&lt;li data-section-
 id=&quot;1hi63o&quot; data-start=&quot;2546&quot; data-end=&quot;2571&quot;&gt;Response sanitization&lt;/li&gt;\n
 &lt;li data-section-id=&quot;1n5xu5u&quot; data-start=&quot;2574&quot; data-end=&quot;2595&quot;&gt;Schema val
 idation&lt;/li&gt;\n&lt;li data-section-id=&quot;1n2kqtv&quot; data-start=&quot;2598&quot; data-end=&quot;26
 19&quot;&gt;Context isolation&lt;/li&gt;\n&lt;/ul&gt;\n&lt;/li&gt;\n&lt;li data-section-id=&quot;v1886t&quot; dat
 a-start=&quot;2621&quot; data-end=&quot;2699&quot;&gt;&lt;strong data-start=&quot;2623&quot; data-end=&quot;2647&quot;&gt;R
 eal-Time Evaluation&lt;/strong&gt;&lt;br data-start=&quot;2647&quot; data-end=&quot;2650&quot;&gt;Prevent 
 attacks without impacting performance&lt;/li&gt;\n&lt;/ul&gt;\n&lt;p data-start=&quot;2701&quot; da
 ta-end=&quot;2783&quot;&gt;👉 &lt;em data-start=&quot;2704&quot; data-end=&quot;2783&quot;&gt;Takeaway: Practic
 al strategies to secure AI agents in enterprise environments&lt;/em&gt;&lt;/p&gt;\n&lt;hr
  data-start=&quot;2785&quot; data-end=&quot;2788&quot;&gt;\n&lt;h2 data-section-id=&quot;14d6kv4&quot; data-st
 art=&quot;2790&quot; data-end=&quot;2812&quot;&gt;About the Organizer&lt;/h2&gt;\n&lt;p data-start=&quot;2814&quot; 
 data-end=&quot;3002&quot;&gt;&lt;strong data-start=&quot;2814&quot; data-end=&quot;2833&quot;&gt;Pradyumna Kodgi&lt;
 /strong&gt;&lt;br data-start=&quot;2833&quot; data-end=&quot;2836&quot;&gt;Principal Product Manager | 
 Oracle Health &amp;amp\; AI&lt;br data-start=&quot;2882&quot; data-end=&quot;2885&quot;&gt;IEEE Senior M
 ember | Vice Chair\, IEEE EMBS &amp;ndash\; Orange County&lt;br data-start=&quot;2943&quot;
  data-end=&quot;2946&quot;&gt;Member\, IEEE AI Agentic Systems &amp;amp\; AI Policy Committ
 ees&lt;/p&gt;\n&lt;p data-start=&quot;3004&quot; data-end=&quot;3073&quot;&gt;📍 California\, USA&lt;br dat
 a-start=&quot;3022&quot; data-end=&quot;3025&quot;&gt;📧 &lt;a class=&quot;decorated-link cursor-pointe
 r&quot; rel=&quot;noopener&quot; data-start=&quot;3028&quot; data-end=&quot;3043&quot;&gt;pkodgi@ieee.org&lt;/a&gt;&lt;br
  data-start=&quot;3043&quot; data-end=&quot;3046&quot;&gt;🔗 linkedin.com/in/pkodgi&lt;/p&gt;&lt;br /&gt;&lt;b
 r /&gt;Agenda: &lt;br /&gt;&lt;h2 data-section-id=&quot;1cbav6o&quot; data-start=&quot;138&quot; data-end=
 &quot;187&quot;&gt;&lt;span role=&quot;text&quot;&gt;&lt;strong data-start=&quot;141&quot; data-end=&quot;187&quot;&gt;Securing A
 I: From Innovation to Resilience&lt;/strong&gt;&lt;/span&gt;&lt;/h2&gt;\n&lt;p data-start=&quot;189&quot;
  data-end=&quot;435&quot;&gt;AI is rapidly transforming how we build intelligent system
 s&amp;mdash\;but as capabilities grow\, so do &lt;strong data-start=&quot;281&quot; data-en
 d=&quot;299&quot;&gt;security risks&lt;/strong&gt;. From LLM-powered agents to tool-integrate
 d architectures\, the question is no longer just &lt;em data-start=&quot;390&quot; data
 -end=&quot;406&quot;&gt;what AI can do&lt;/em&gt;&amp;mdash\;but &lt;strong data-start=&quot;411&quot; data-en
 d=&quot;435&quot;&gt;how do we secure it?&lt;/strong&gt;&lt;/p&gt;\n&lt;p data-start=&quot;437&quot; data-end=&quot;6
 26&quot;&gt;In this interactive session\, we cut through the noise and break down 
 &lt;strong data-start=&quot;506&quot; data-end=&quot;552&quot;&gt;AI security in practical\, real-wo
 rld terms&lt;/strong&gt;&amp;mdash\;so you can understand not just the risks\, but h
 ow to defend against them.&lt;/p&gt;\n&lt;hr data-start=&quot;628&quot; data-end=&quot;631&quot;&gt;\n&lt;h3 
 data-section-id=&quot;138f40v&quot; data-start=&quot;633&quot; data-end=&quot;663&quot;&gt;&lt;span role=&quot;text
 &quot;&gt;🔍 &lt;strong data-start=&quot;640&quot; data-end=&quot;663&quot;&gt;What You&amp;rsquo\;ll Explore&lt;
 /strong&gt;&lt;/span&gt;&lt;/h3&gt;\n&lt;ul data-start=&quot;665&quot; data-end=&quot;932&quot;&gt;\n&lt;li data-secti
 on-id=&quot;13mqes9&quot; data-start=&quot;665&quot; data-end=&quot;740&quot;&gt;How modern AI systems (LLM
 s\, agents\, MCP) introduce new attack surfaces&lt;/li&gt;\n&lt;li data-section-id=
 &quot;137jdro&quot; data-start=&quot;741&quot; data-end=&quot;811&quot;&gt;The shift from traditional secur
 ity to &lt;strong data-start=&quot;782&quot; data-end=&quot;809&quot;&gt;AI-driven threat models&lt;/st
 rong&gt;&lt;/li&gt;\n&lt;li data-section-id=&quot;85e0fd&quot; data-start=&quot;812&quot; data-end=&quot;871&quot;&gt;K
 ey security concepts&amp;mdash\;explained clearly and practically&lt;/li&gt;\n&lt;li da
 ta-section-id=&quot;oenlum&quot; data-start=&quot;872&quot; data-end=&quot;932&quot;&gt;Real-world attack s
 cenarios and emerging threat patterns&lt;/li&gt;\n&lt;/ul&gt;\n&lt;hr data-start=&quot;934&quot; da
 ta-end=&quot;937&quot;&gt;\n&lt;h3 data-section-id=&quot;13oed0c&quot; data-start=&quot;939&quot; data-end=&quot;98
 3&quot;&gt;&lt;span role=&quot;text&quot;&gt;💡 &lt;strong data-start=&quot;946&quot; data-end=&quot;983&quot;&gt;What Mak
 es This Session Different&lt;/strong&gt;&lt;/span&gt;&lt;/h3&gt;\n&lt;p data-start=&quot;985&quot; data-e
 nd=&quot;1069&quot;&gt;This isn&amp;rsquo\;t just theory&amp;mdash\;you&amp;rsquo\;ll see &lt;strong d
 ata-start=&quot;1019&quot; data-end=&quot;1068&quot;&gt;AI systems under attack and defense in ac
 tion&lt;/strong&gt;.&lt;/p&gt;\n&lt;p data-start=&quot;1071&quot; data-end=&quot;1236&quot;&gt;Through a &lt;strong
  data-start=&quot;1081&quot; data-end=&quot;1115&quot;&gt;live\, end-to-end demonstration&lt;/strong
 &gt;\, we&amp;rsquo\;ll show how AI agents can be manipulated&amp;mdash\;and how laye
 red security approaches can prevent these attacks in real time.&lt;/p&gt;\n&lt;hr d
 ata-start=&quot;1238&quot; data-end=&quot;1241&quot;&gt;\n&lt;h3 data-section-id=&quot;10uroyp&quot; data-star
 t=&quot;1243&quot; data-end=&quot;1274&quot;&gt;&lt;span role=&quot;text&quot;&gt;🛠️ &lt;strong data-start=&quot;125
 1&quot; data-end=&quot;1274&quot;&gt;Practical Takeaways&lt;/strong&gt;&lt;/span&gt;&lt;/h3&gt;\n&lt;p data-start
 =&quot;1276&quot; data-end=&quot;1376&quot;&gt;You&amp;rsquo\;ll walk away with &lt;strong data-start=&quot;1
 298&quot; data-end=&quot;1338&quot;&gt;actionable strategies and frameworks&lt;/strong&gt; you can
  apply immediately\, including:&lt;/p&gt;\n&lt;ul data-start=&quot;1377&quot; data-end=&quot;1540&quot;
 &gt;\n&lt;li data-section-id=&quot;xcdr3y&quot; data-start=&quot;1377&quot; data-end=&quot;1431&quot;&gt;Securing
  AI agents interacting with external tools&lt;/li&gt;\n&lt;li data-section-id=&quot;13wk
 hll&quot; data-start=&quot;1432&quot; data-end=&quot;1478&quot;&gt;Validating and sanitizing untrusted
  inputs&lt;/li&gt;\n&lt;li data-section-id=&quot;1ispc52&quot; data-start=&quot;1479&quot; data-end=&quot;15
 40&quot;&gt;Designing &lt;strong data-start=&quot;1491&quot; data-end=&quot;1511&quot;&gt;trust boundaries&lt;/
 strong&gt; in AI-driven architectures&lt;/li&gt;\n&lt;/ul&gt;\n&lt;hr data-start=&quot;1542&quot; data
 -end=&quot;1545&quot;&gt;\n&lt;h3 data-section-id=&quot;mtmo82&quot; data-start=&quot;1547&quot; data-end=&quot;157
 5&quot;&gt;&lt;span role=&quot;text&quot;&gt;🎯 &lt;strong data-start=&quot;1554&quot; data-end=&quot;1575&quot;&gt;Who Sh
 ould Attend&lt;/strong&gt;&lt;/span&gt;&lt;/h3&gt;\n&lt;ul data-start=&quot;1577&quot; data-end=&quot;1824&quot;&gt;\n
 &lt;li data-section-id=&quot;1juh51w&quot; data-start=&quot;1577&quot; data-end=&quot;1642&quot;&gt;Security p
 rofessionals and architects working with AI systems&lt;/li&gt;\n&lt;li data-section
 -id=&quot;ptvlgl&quot; data-start=&quot;1643&quot; data-end=&quot;1706&quot;&gt;Engineers and developers bu
 ilding AI/LLM-based applications&lt;/li&gt;\n&lt;li data-section-id=&quot;3l4mav&quot; data-s
 tart=&quot;1707&quot; data-end=&quot;1759&quot;&gt;Product managers and leaders driving AI adopti
 on&lt;/li&gt;\n&lt;li data-section-id=&quot;nc6kx0&quot; data-start=&quot;1760&quot; data-end=&quot;1824&quot;&gt;An
 yone interested in understanding &lt;strong data-start=&quot;1797&quot; data-end=&quot;1822&quot;
 &gt;AI risks and defenses&lt;/strong&gt;&lt;/li&gt;\n&lt;/ul&gt;\n&lt;hr data-start=&quot;1826&quot; data-en
 d=&quot;1829&quot;&gt;\n&lt;h3 data-section-id=&quot;11kjytf&quot; data-start=&quot;1831&quot; data-end=&quot;1867&quot;
 &gt;&lt;span role=&quot;text&quot;&gt;✨ &lt;strong data-start=&quot;1837&quot; data-end=&quot;1867&quot;&gt;What You&amp;
 rsquo\;ll Walk Away With&lt;/strong&gt;&lt;/span&gt;&lt;/h3&gt;\n&lt;ul data-start=&quot;1869&quot; data-
 end=&quot;2069&quot;&gt;\n&lt;li data-section-id=&quot;180pwx7&quot; data-start=&quot;1869&quot; data-end=&quot;192
 8&quot;&gt;A clear understanding of &lt;strong data-start=&quot;1896&quot; data-end=&quot;1926&quot;&gt;emer
 ging AI security risks&lt;/strong&gt;&lt;/li&gt;\n&lt;li data-section-id=&quot;1qjuuw3&quot; data-s
 tart=&quot;1929&quot; data-end=&quot;1995&quot;&gt;Practical knowledge of how to &lt;strong data-sta
 rt=&quot;1961&quot; data-end=&quot;1993&quot;&gt;secure AI agents and systems&lt;/strong&gt;&lt;/li&gt;\n&lt;li 
 data-section-id=&quot;h35u0m&quot; data-start=&quot;1996&quot; data-end=&quot;2069&quot;&gt;Real-world insi
 ghts into &lt;strong data-start=&quot;2023&quot; data-end=&quot;2067&quot;&gt;attack prevention and 
 defense strategies&lt;/strong&gt;&lt;/li&gt;\n&lt;/ul&gt;\n&lt;hr data-start=&quot;2071&quot; data-end=&quot;2
 074&quot;&gt;\n&lt;p data-start=&quot;2076&quot; data-end=&quot;2296&quot;&gt;As AI systems become more auto
 nomous and integrated into enterprise workflows\, &lt;strong data-start=&quot;2155
 &quot; data-end=&quot;2201&quot;&gt;security becomes foundational&amp;mdash\;not optional&lt;/stron
 g&gt;. This session will equip you with the mindset and tools to &lt;strong data
 -start=&quot;2261&quot; data-end=&quot;2296&quot;&gt;build AI systems you can trust.&lt;/strong&gt;&lt;/p&gt;
END:VEVENT
END:VCALENDAR