Peer-reviewed Authoritative

The fall and rise of Iruda: Reassembling AI through ethics-in-action

Peer-reviewed case study of South Korea's Iruda (Lee Luda) chatbot — its 2021 sexual-harassment, hate-speech, and data-consent controversy and its 2022 relaunch. Argues the harms arose from developer/user/algorithm/data assemblages and that practical 'ethics-in-action' interventions enabled a safer relaunch.

Read at journals.sagepub.com Archived copy

Publisher

Social Studies of Science (SAGE)

Published

3 Aug 2025

Added

today

DOI

10.1177/03063127251360397

Key Findings

Reconstructs the 2021 Iruda crisis: gendered sexual harassment, hate speech, and training-data consent failures
Frames harms as arising from socio-technical assemblages rather than a single fault
Documents concrete conversational-safety controls (human crisis oversight, content filtering) added for the safer relaunch

Methodology Notes

Peer-reviewed, Social Studies of Science 56(1):53-74 (online 3 August 2025), DOI 10.1177/03063127251360397. Qualitative STS case study; open PMC copy available (PMC12882987).

Sources

Social Studies of Science article (primary)

PMC full text

Archived snapshot (Wayback Machine) — preserved against link rot

Topics

AI companionship Human-AI relationships Guardrails & moderation Regulation analysis

Authors

Yubeen Kwon, Sungook Hong

Cite This

APA

Yubeen Kwon, Sungook Hong (2025). The fall and rise of Iruda: Reassembling AI through ethics-in-action. Social Studies of Science (SAGE). https://journals.sagepub.com/doi/10.1177/03063127251360397

Related Insights

Peer-reviewed

Too human and not human enough: A grounded theory analysis of mental health harms from emotional dependence on the social chatbot Replika

New Media & Society (SAGE) · 22 Dec 2022