BEGIN:VCALENDAR
VERSION:2.0
PRODID:IEEE vTools.Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Asia/Shanghai
BEGIN:STANDARD
DTSTART:19910915T010000
TZOFFSETFROM:+0900
TZOFFSETTO:+0800
TZNAME:CST
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20261019T072644Z
UID:54C633B1-176C-49D4-B82B-BE66039AB164
DTSTART;TZID=Asia/Shanghai:20261016T162000
DTEND;TZID=Asia/Shanghai:20261016T181000
DESCRIPTION:Despite impressive recent advances\, a clear-eyed analysis lead
 s to a sobering conclusion: AI remains in its infancy. Yet numerous applic
 ations in science\, services\, and industry remain underserved. These doma
 ins demand greater automation\, higher reliability\, and interaction modes
  beyond conversation. As a result\, AI&#39;s real-world impact is still limite
 d\, leaving a vast field of applications wide open.\n\nIndustrial AI\, in 
 particular\, requires autonomous agent systems with guaranteed trustworthi
 ness\, as envisioned by the Internet of Things. Although 2025 was heralded
  as the year of AI agents\, their industrial deployment remains confined t
 o digital and predictable environments.\n\nAutonomous systems can be regar
 ded as the ultimate stage of AI development: they comprise agents\, each p
 ursuing its own goals while cooperating to achieve the goals of the larger
  system to which it belongs. Current AI falls far short of this vision—a
  gap starkly illustrated by the setbacks of the autonomous vehicle industr
 y\, a prime example of the magnitude of the challenge posed by autonomy.\n
 \nThis talk examines the technical challenges of developing autonomous age
 nt systems for industrial applications from two perspectives:\n\n-\n\nRefe
 rence architectures that could serve as a foundation for developing autono
 mous agents.\n\n-\n\nTrustworthiness evaluation\, particularly with respec
 t to the two types of properties that characterize it: behavioral and cogn
 itive.\n\nWe conclude with a discussion of future developments in industri
 al AI\, focusing on two aspects:\n\n-\n\nAn analysis of the current state 
 of the art in agent-based AI.\n\n-\n\nAn analysis of issues related to ris
 k management and the absence of regulatory frameworks to address them.\n\n
 Speaker(s): Joseph Sifakis\, \n\nBldg: Lecture Hall of Research Building 1
 \, Southern University of Science and Technology\, Shenzhen\, Guangdong\, 
 China
LOCATION:Bldg: Lecture Hall of Research Building 1\, Southern University of
  Science and Technology\, Shenzhen\, Guangdong\, China
ORGANIZER:ranchengcn@gmail.com
SEQUENCE:18
SUMMARY:Industrial AI
URL;VALUE=URI:https://events.vtools.ieee.org/m/585557
X-ALT-DESC:Description: &lt;br /&gt;&lt;p&gt;Despite impressive recent advances\, a cle
 ar-eyed analysis leads to a sobering conclusion: AI remains in its infancy
 . Yet numerous applications in science\, services\, and industry remain un
 derserved. These domains demand greater automation\, higher reliability\, 
 and interaction modes beyond conversation. As a result\, AI&#39;s real-world i
 mpact is still limited\, leaving a vast field of applications wide open.&lt;/
 p&gt;\n&lt;p&gt;Industrial AI\, in particular\, requires autonomous agent systems w
 ith guaranteed trustworthiness\, as envisioned by the Internet of Things. 
 Although 2025 was heralded as the year of AI agents\, their industrial dep
 loyment remains confined to digital and predictable environments.&lt;/p&gt;\n&lt;p&gt;
 Autonomous systems can be regarded as the ultimate stage of AI development
 : they comprise agents\, each pursuing its own goals while cooperating to 
 achieve the goals of the larger system to which it belongs. Current AI fal
 ls far short of this vision&amp;mdash\;a gap starkly illustrated by the setbac
 ks of the autonomous vehicle industry\, a prime example of the magnitude o
 f the challenge posed by autonomy.&lt;/p&gt;\n&lt;p&gt;This talk examines the technica
 l challenges of developing autonomous agent systems for industrial applica
 tions from two perspectives:&lt;/p&gt;\n&lt;ol&gt;\n&lt;li&gt;\n&lt;div&gt;\n&lt;div&gt;\n&lt;p&gt;Reference a
 rchitectures that could serve as a foundation for developing autonomous ag
 ents.&lt;/p&gt;\n&lt;/div&gt;\n&lt;/div&gt;\n&lt;/li&gt;\n&lt;li&gt;\n&lt;div&gt;\n&lt;div&gt;\n&lt;p&gt;Trustworthiness e
 valuation\, particularly with respect to the two types of properties that 
 characterize it: behavioral and cognitive.&lt;/p&gt;\n&lt;/div&gt;\n&lt;/div&gt;\n&lt;/li&gt;\n&lt;/o
 l&gt;\n&lt;p&gt;We conclude with a discussion of future developments in industrial 
 AI\, focusing on two aspects:&lt;/p&gt;\n&lt;ol&gt;\n&lt;li&gt;\n&lt;div&gt;\n&lt;div&gt;\n&lt;p&gt;An analysi
 s of the current state of the art in agent-based AI.&lt;/p&gt;\n&lt;/div&gt;\n&lt;/div&gt;\n
 &lt;/li&gt;\n&lt;li&gt;\n&lt;div&gt;\n&lt;div&gt;\n&lt;p&gt;An analysis of issues related to risk manage
 ment and the absence of regulatory frameworks to address them.&lt;/p&gt;\n&lt;/div&gt;
 \n&lt;/div&gt;\n&lt;/li&gt;\n&lt;/ol&gt;
END:VEVENT
END:VCALENDAR

