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:20260222T003643Z
UID:FF477028-4F70-42A2-A077-FE8FEF3EA0C0
DTSTART;TZID=America/Los_Angeles:20260414T190000
DTEND;TZID=America/Los_Angeles:20260414T210000
DESCRIPTION:This is a hybrid in-person and online event. Pre-registration i
 s required for either.\n\nConversational AI systems today speak with remar
 kable confidence\, often giving the impression of understanding and reason
 ing. However\, teams deploying these systems often quickly encounter famil
 iar problems: drift\, hallucinations\, contradictory answers\, and convers
 ations that quietly lose their original purpose. Why do systems that seem 
 so capable end up behaving so unpredictably?\n\nIn this talk\, Elena Gostr
 er will examine these behaviors from a practical\, product-engineering per
 spective. Rather than exploring model internals\, this talk will focus on 
 what actually happens when humans interact with probabilistic language mod
 els – and why traditional software assumptions fail in conversational se
 ttings.\n\nElena will also discuss what architectural patterns teams are a
 dopting to keep in control\, and she’ll highlight why combining generati
 ve AI with explicit structure\, state\, and constraints is becoming essent
 ial. Attendees will leave with a clearer understanding of why conversation
 al AI breaks\, what helps it behave more reliably\, and how to think diffe
 rently about designing human-AI interactions.\n\nSpeaker(s): Elena Gostrer
 \, \n\n925 Thompson Place\, Sunnyvale\, California\, United States\, 94085
 \, Virtual: https://events.vtools.ieee.org/m/541387
LOCATION:925 Thompson Place\, Sunnyvale\, California\, United States\, 9408
 5\, Virtual: https://events.vtools.ieee.org/m/541387
ORGANIZER:dmsnyder@ieee.org
SEQUENCE:16
SUMMARY:Conversational AI: Practical Challenges in Talking to People
URL;VALUE=URI:https://events.vtools.ieee.org/m/541387
X-ALT-DESC:Description: <br /><p><strong>This is a hybrid in-person and onl
 ine event. Pre-registration is required for either.</strong></p>\n<p>Conve
 rsational AI systems today speak with remarkable confidence\, often giving
  the impression of understanding and reasoning. However\, teams deploying 
 these systems often quickly encounter familiar problems: drift\, hallucina
 tions\, contradictory answers\, and conversations that quietly lose their 
 original purpose. Why do systems that seem so capable end up behaving so u
 npredictably?</p>\n<p>In this talk\, Elena Gostrer will examine these beha
 viors from a practical\, product-engineering perspective. Rather than expl
 oring model internals\, this talk will focus on what actually happens when
  humans interact with probabilistic language models &ndash\; and why tradi
 tional software assumptions fail in conversational settings.</p>\n<p>Elena
  will also discuss what architectural patterns teams are adopting to keep 
 in control\, and she&rsquo\;ll highlight why combining generative AI with 
 explicit structure\, state\, and constraints is becoming essential. Attend
 ees will leave with a clearer understanding of why conversational AI break
 s\, what helps it behave more reliably\, and how to think differently abou
 t designing human-AI interactions.</p>
END:VEVENT
END:VCALENDAR