The ethics of advanced ai assistants

I Gabriel, A Manzini, G Keeling, LA Hendricks… - arXiv preprint arXiv …, 2024 - arxiv.org
This paper focuses on the opportunities and the ethical and societal risks posed by
advanced AI assistants. We define advanced AI assistants as artificial agents with natural …

The PRISM Alignment Project: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of …

HR Kirk, A Whitefield, P Röttger, A Bean… - arXiv preprint arXiv …, 2024 - arxiv.org
Human feedback plays a central role in the alignment of Large Language Models (LLMs).
However, open questions remain about the methods (how), domains (where), people (who) …

AI ethics as a complex and multifaceted challenge: decoding educators' AI ethics alignment through the lens of activity theory

J Kamali, MF Alpat, A Bozkurt - International Journal of Educational …, 2024 - Springer
This study explores university educators' perspectives on their alignment with artificial
intelligence (AI) ethics, considering activity theory (AT), which forms the theoretical …

Beyond static AI evaluations: advancing human interaction evaluations for LLM harms and risks

L Ibrahim, S Huang, L Ahmad, M Anderljung - arXiv preprint arXiv …, 2024 - arxiv.org
Model evaluations are central to understanding the safety, risks, and societal impacts of AI
systems. While most real-world AI applications involve human-AI interaction, most current …

Participation in the age of foundation models

H Suresh, E Tseng, M Young, M Gray… - The 2024 ACM …, 2024 - dl.acm.org
Growing interest and investment in the capabilities of foundation models has positioned
such systems to impact a wide array of services, from banking to healthcare. Alongside …

ValueScope: Unveiling Implicit Norms and Values via Return Potential Model of Social Interactions

CY Park, SS Li, H Jung, S Volkova, T Mitra… - arXiv preprint arXiv …, 2024 - arxiv.org
This study introduces ValueScope, a framework leveraging language models to quantify
social norms and values within online communities, grounded in social science perspectives …

An Ellulian analysis of propaganda in the context of generative AI

X Bi, X Su, X Liu - Ethics and Information Technology, 2024 - Springer
The application of generative artificial intelligence (GenAI) technologies in the field of
propaganda influences information creation, dissemination, and reception, and introduces …

Beyond the Binary: Capturing Diverse Preferences With Reward Regularization

V Padmakumar, C Jin, HR Kirk, H He - arXiv preprint arXiv:2412.03822, 2024 - arxiv.org
Large language models (LLMs) are increasingly deployed via public-facing interfaces to
interact with millions of users, each with diverse preferences. Despite this, preference tuning …

Applying a community‐engaged participatory machine learning model

EN Asabor, K Aneni, S Weerakoon… - American Journal of …, 2024 - Wiley Online Library
Although predictive algorithms have been described as the definitive solution to bias in
health care, machine learning techniques may also propagate existing health inequities …

Participation versus scale: Tensions in the practical demands on participatory AI

M Young, U Ehsan, R Singh, E Tafesse, M Gilman… - First Monday, 2024 - firstmonday.org
Ongoing calls from academic and civil society groups and regulatory demands for the
central role of affected communities in development, evaluation, and deployment of artificial …