讲座介绍：

Generative AI (GenAI) platforms have transformed digital creationbut introduced new challenges for information governance. Despite activemoderation, these platforms still surface explicit or boundary-pushing outputsthat pose ethical and legal risks. Existing safeguards assess prompts oroutputs in isolation, overlooking how users adapt across iterative interactionswith GenAI. Consequently, current governance mechanisms fail to capture howeven rare exposures to explicit outputs reshape user behavior over time,creating an unrecognized behavioral source of governance risk. We address thisgap by developing a trajectory-aware governance framework that links user-levelbehavioral dynamics with platform-level moderation design. Using a large-scaledataset from a leading text-to-image platform and a hidden Markov model toinfer latent engagement states and transitions, we show that exposure toexplicit content drives escalation and persistence in higher-intensityengagement. The effect is more substantial for active seekers, weaker for users with public profiles, and attenuates withtenure, consistent with curiosity-driven behavioral theory. At the platformlevel, we introduce a platform-tolerance metric, the mean explicitness ofgenerated outputs, to quantify platform tolerance and simulate policy outcomes.Results reveal a measurable safety–engagement trade-off: a tighter platformreduces exposure but suppresses participation and topic diversity, suggestingan optimal tolerance level that balances safety and engagement. This studyextends information governance from technical control to behavioral adaptation,demonstrating that effective governance in GenAI requires dynamic,behavior-aware governance mechanisms. It contributes a new behavioral sourcefor governance risk (exposure-driven escalation), a quantifiable governanceconstruct (platform tolerance), and an integrated modeling framework forassessing how policy design shapes safety and user engagement. For platformmanagers, it provides actionable guidance on developing governance policiesthat strike a balance between user safety and the creative potential of GenAI.

主讲人介绍：

Lu Huang，现任美国宾夕法尼亚州立大学斯米尔商学院Assistant Clinical Professor，获美国康涅狄格大学定量建模博士学位。其研究成果发表于多个商学与经济学领域重要期刊，包括Production and Operations Management、Journal of Interactive Marketing、Journal of Public Policy & Marketing以International Journal of Industrial Organization 等，主要研究聚焦于机器学习与动态结构计量经济模型，为管理实践与商业决策提供方法支持与实证依据。