Keynote Speaker 1: Dr. Fabio Casati

Title: Understanding the Behavior and Quality of AI Services: Obvious Things that Need to Be Said

Abstract: AI-powered services, representative of 'Software 3.0', follow a development paradigm where progress is measured by quality improvement rather than new features. This creates a core dependency on an 'iterate-evaluate-improve' loop analogous to training AI models. Consequently, the accuracy of the evaluation process is the cornerstone of the entire system: precise evaluations drive effective improvement, while noisy evaluations severely hinder progress. A central challenge, however, is that while evaluation is foundational and AI simplifies development, building a reliable evaluation process is exceptionally difficult. This difficulty is often underestimated by teams lacking an AI-first culture. This talk will illuminate the brittleness of evaluation in AI services and explore how to strengthen it—primarily by correctly defining the problems we aim to solve—to accelerate service improvement. The presentation will not introduce novel knowledge but will instead synthesize established concepts to provide a clear framework for building more reliable evaluation processes.

Biography: Dr. Fabio Casati is Professor at the University of Trento and principal AI architect for ServiceNow. Fabio focuses on designing, architecting and deploying AI-powered workflows for enterprise customers. On the research side, he is working on evaluations and governance of AI systems and on AI systems that serve needs of individuals and subjective point of views.

Keynote Speaker 2: Dr. Yi Xie

Title: Research and development practice of Huawei Celia Agent and HarmonyOS Agent Framework

Abstract: At a time when AI technology is revolutionizing the paradigm of human-device interaction, the HarmonyOS Intelligent Agent, implemented in Huawei's HarmonyOS 6.0, is driving a transition in operating systems from 'passively responding to commands' to 'actively understanding and making decisions.' This presentation will focus on this core technology, introducing the HarmonyOS Intelligent Agent Framework (HMAF) and related technologies and scenarios of Huawei's Xiaoyi, while exploring how Agents are reshaping device experience and ecosystem dynamics.

Biography: Yi Xie, Ph.D. in Engineering from Zhejiang University, was a visiting scholar at McGill University. He currently serves as the Head of AI Agent R&D at Huawei Consumer Business Group, leading the technology and product development related to Xiaoyi and HarmonyOS Intelligent Agents. With over 30 filed patents in artificial intelligence, his collaborative research has been published at top-tier AI conferences such as AAAI and NeurIPS. He has spearheaded the development of key HarmonyOS capabilities including the Xiaoyi Agent Platform and Xiaoyi Memory. He also led the authorship of the 'HarmonyOS Intelligent Agent Framework White Paper,' which was publicly released at Huawei HDC 2025.

Keynote Speaker 3: Dr. Zibin Zheng

Title: Trustworthy Services: From Trustworthy Blockchains to Trustworthy LLMs

Abstract: In recent years, digital services such as blockchain-as-a-service and AI-as-a-service have advanced rapidly. However, digital services also entail trustworthiness risks, and ensuring the trustworthiness of digital services has become an urgent issue. This talk will present the trustworthiness challenges of blockchain services and LLM (Large Language Model) services across multiple dimensions, and then introduce related technical research and industry enablement case studies from the School of Software Engineering, Sun Yat-sen University.

Biography: Dr. Zibin Zheng is the Dean of the School of Software Engineering, and the Vice Dean of the Artificial Intelligence Institute at Sun Yat-sen University. He is an IEEE Fellow, an IET Fellow, and an ACM Distinguished Scientist. He also serves as the Deputy Director of the National Digital Home Engineering Technology Research Center and the Director of the Guangdong Provincial Blockchain Engineering Technology Research Center. His research interests include service computing, trustworthy large language models, blockchain, and software reliability. According to Google Scholar, his publications have received over 50,000 citations, with an H-index of 100. His honors include the Second Prize of Natural Science Award from the Ministry of Education, the Second Prize of Wenjun Wu AI Natural Science Award, and the ACM SIGSOFT Distinguished Paper Award.