Navigation auf uzh.ch
Navigation auf uzh.ch
The theme of the series «Philosophy of AI» is to present various debates in the philosophy of AI. The talks will be offered by experts coming from several disciplines around AI and digital innovation, and the aim of the series is generally to foster cross-pollination and interaction between different fields. This includes (not exhaustively) philosophy, engineering, information systems, informatics, law, political sciences, sociology.
4. Session: «Addressing Social Misattributions of Large Language Models»
The talk is based on a work co-authored with Alessandro Facchini (IDSIA USI-SUPSI) and Andrea Ferrario (UZH). Human-centered explainable AI (HCXAI) advocates for the integration of social aspects into AI explanations. Central to the HCXAI discourse is the Social Transparency (ST) framework, which aims to make the socio-organizational context of AI systems accessible to their users. In this work, we suggest extending the ST framework to address the risks of social misattributions in Large Language Models (LLMs), particularly in sensitive areas like mental health. In fact LLMs, which are remarkably capable of simulating roles and personas, may lead to mismatches between designers’ intentions and users’ perceptions of social attributes, risking to promote emotional manipulation and dangerous behaviors, cases of epistemic injustice, and unwarranted trust. To address these issues, we propose enhancing the ST framework with a fifth «W-question» to clarify the specific social attributions assigned to LLMs by its designers and users. This addition aims to bridge the gap between LLM capabilities and user perceptions, promoting the ethically responsible development and use of LLM-based technology.
When / Wann: May 28, 2024, 17:00 - 19:30 (incl. Apéro)
Where / Wo: Digital Library Space, Rämistrasse 69, 8001 Zurich
Language / Sprache: English
Registration / Anmeldung: Link
Organisation: DSI Community Ethics
