BIMSA >
BIMSA Digital Economy Lab Seminar
BIMSA Digital Economy Lab Seminar
From Pixels to Agents: Building Trustworthy and Secure AI
From Pixels to Agents: Building Trustworthy and Secure AI
Organizers
Johansson Anders
,
Manyao Deng
,
Ruize Gao
,
Liyan Han
,
Zhen Li
,
Jin Liu
,
Fei Long
,
Dongbo Shi
,
Ke Tang
,
Xing Yan
,
Qi Zhang
Speaker
Time
Thursday, November 5, 2026 3:00 PM - 4:00 PM
Venue
A3-2-303
Online
Zoom 242 742 6089
(BIMSA)
Abstract
As AI becomes part of scientific research and everyday decision-making, strong performance alone is not enough: systems must also resist manipulation and remain reliable when deployed. This talk introduces research on trustworthy and secure AI through three connected questions: How can AI be misled? How can it be protected? And how can users check that a deployed system has not been altered? Accessible examples spanning image recognition, conversational assistants, and AI systems that search for information and carry out tasks illustrate both emerging risks and practical approaches to addressing them. The aim is to connect these challenges with problems in other disciplines and explore opportunities for collaboration on realistic applications, shared evaluation methods, and the mathematical foundations of trustworthy AI.
Speaker Intro
Bin Zhu worked at Microsoft from December 2001 through June 2026, including about 22 years at Microsoft Research Asia and later in the M365 Core and Bing Core Search product teams. He was a Principal Researcher at Microsoft Research Asia and in the product groups. His research spans trustworthy AI, AI security, AI applications in security, trustworthy computing, privacy-preserving computation, authentication, CAPTCHA, web systems, and multimedia security. His recent work includes AI security and privacy research, particularly LLM and agent safety, model tamper detection and IP protection, adversarial robustness, and privacy-preserving computation, alongside AI-assistant reasoning, deep search, evaluation, prompt optimization, document understanding, and multi-vector retrieval. His research has been transferred into Microsoft products and services for ransomware defense, malicious PowerShell script detection, phishing and fraud detection, CAPTCHA, Bing index compression, and mobile integrity checking. He has published 125+ papers, holds 58 issued U.S. patents, and has long been involved in student mentoring and research collaborations with universities in China, the U.S., and the U.K.