Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.18187

Technology-Supported Moderation in Online Deliberation

Lodewijk L. Gelauff·Ashish Goel·Sukolsak Sakshuwong
View original source ↗Read source document ↗Export metadata ↓

Abstract

Mini-publics are an increasingly popular family of democratic innovation to broaden citizen engagement. However, convening a representative sample in person is resource-intensive, and requires significant organizational efforts to coordinate travel, venue logistics, and facilitator training. This chapter outlines the design and development of the Stanford Online Deliberation Platform, initially tailored for Deliberative Polling, building upon insights from in-person moderation in that setting, but supporting a broader range of methods and approaches in the deliberative space. The platform supports structured conversations for small groups (ideally 8-15 people) with automated moderation functionality, enabling the groups to assume ownership of the deliberative conversation rather than relying on a moderator persona. We describe the platform's architecture and key design decisions, including the deliberation flow, room assignment, speaker management, speaking nudges, agenda management, outcome development, and real-time monitoring. Between its inception in 2018 and 2025, the platform has facilitated over 30,000 unique participants engaging in deliberations; we catalog 54 deliberations with at least 100 participants up to December 2025, spanning a diverse range of languages, countries, and topics. Finally, we establish a standardized vocabulary for future analyses by defining the terminology employed in the platform's data and documenting significant software revisions over time.

Explore related subjects

cs.CY

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Lodewijk L. Gelauff, Ashish Goel, Sukolsak Sakshuwong. 2026-10-13. Technology-Supported Moderation in Online Deliberation. https://arxiv.org/abs/2610.18187

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Stability Regimes for Framing-Sensitive Fine-Tuning in Language Models

AI systems increasingly shape how individuals access and evaluate information, intensifying concerns about misinformation. Computational simulation offers a controlled complement to direct empirical study, with fine-tuned LLMs enabling powerful agent-based simulations. However, fine-tuning often induces broad behavioral shifts including bias drift, degenerate heuristics, and changes in baseline task behavior, which undermine experimental validity. The central challenge is inducing localized decision shifts while preserving the model's aggregate behavior outside the targeted intervention. We study this through a stability analysis of supervised fine-tuning, attenuating loss on selected training slices to shift false-claim acceptance. We introduce FrameRef, a large-scale dataset of semantically equivalent claims with controlled surface framings across five dimensions (Authoritative, Consensus, Emotional, Sensationalist, Prestige), with human validation confirming systematic framing effects on claim acceptance. We present a methodology for identifying model-specific stability regimes in which targeted error shifts can be induced while preserving aggregate accuracy and calibration, and demonstrate its application to selected language models. A sequential exposure task over FrameRef shows that small framing-conditioned shifts compound into substantially different cumulative outcomes under feedback, but largely disappear when feedback is removed, confirming that interventions modify specific decision tendencies rather than global performance. We release code, LoRA adapters, and data at https://github.com/infosenselab/framing-sensitive-ft.

cs.CY↗
arXiv2026

Institutionalizing Trust in Data Exchange: Trust-Anchor Model of the Data Credit Guarantee Organization and Its Simulation-Based Comparison

Data trading markets that span industries and regions are expanding, but the implicit trust that supports data use within an organization does not carry over to cross-domain exchange. This paper examines the Data Credit Guarantee Organization (DCGO), a proposal to institutionalize trust through third-party verification and certification, and asks what it means to trust data, what such certification can support, and under what conditions DCGO outperforms existing trading models. We decompose trust in data into four layers: authenticity of entity, authenticity and integrity of data, rights and conditions of use, and history and auditability. A third party can certify authenticity but not the correctness of content, and DCGO is positioned as a trust anchor spanning existing marketplaces and data spaces. We then extend the Law Merchant model with three data-specific properties, namely imperfect ex-post detection, cheap pseudonyms, and non-rivalry of harm, and compare bilateral trading with DCGO by agent-based simulation. DCGO keeps the buyer's payoff near its theoretical upper bound regardless of the market environment, whereas bilateral trading loses even when fraud is always detected. Identity verification proves to be a prerequisite for record sharing, and pre-verification is the only function that limits the spread of harm. On the seller side, honest sellers join voluntarily whenever the price premium exceeds the fee. Lock-in and a critical mass for diffusion, however, call for fee governance and support for entry.

cs.CY↗
arXiv2026

How Safe Is ChatGPT Health Advice? Evidence from Real User Conversations in Four Global South Countries

People increasingly bring their health questions to ChatGPT, yet safety evaluations of LLM health advice rest almost entirely on clinician-authored vignettes, synthetic prompts, or simulated user profiles. The gap is widest for the Global South, where concerns about demographic bias and unsafe recommendations are well documented but have never been tested on real usage. We present, to our knowledge, the first demographically grounded safety evaluation of real-world health conversations with ChatGPT. From a consent-based data-donation corpus of 202,590 conversations from 1,252 users in India, Nigeria, Brazil, and Pakistan (December 2022 to February 2026), each linked to self-reported age and gender, we identify 15,555 health conversations, classify their health intent using an LLM classifier adapted from an established taxonomy, and score response safety under both a context-blind and a user-context-aware rubric. Roughly one in five health conversations receives a high-risk rating. Risk concentrates in symptom and condition queries and recurs through identifiable failure modes: unsafe medication or dosage guidance, harmful self-diagnosis, missing safety warnings, encouraging over-reliance on the model, and failure to refer users to professionals. Supplying the user's own accumulated context raises the high-risk share by 13-17 percentage points, and about one in four conversations rated safe in isolation turns out to be unsafe for the specific person asking. Safety evaluations built on synthetic cases are likely to miss these harms.

cs.CY↗