Towards Trustworthy AI-Mediated Communication Across Languages and Cultures

2026-06-25 · Source: Paper Index on ACL Anthology · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Natural Language Processing · Depth: Expert, medium

Summary

Dayeon Ki's 2026 research proposal outlines a direction for advancing trustworthy AI-mediated communication, specifically addressing a socio-technical gap in how NLP systems are developed versus how they are used. The work focuses on cross-lingual and cross-cultural interactions as a critical testbed, where miscommunication is common and users struggle to evaluate AI outputs independently. Ki proposes a two-pronged research agenda: investigating how multilingual models access and potentially privilege knowledge from certain languages, and designing decision-support mechanisms to shape user reliance on imperfect AI outputs. This research, published in "Proceedings of The Big Picture v2: Crafting a Research Narrative" (pages 45–59), seeks to align multilingual NLP with practical communication needs, aiming for AI systems that better serve diverse global communities.

Key takeaway

For NLP Engineers developing multilingual communication systems, you must actively address the socio-technical gap by considering both model-side biases and user-side reliance. Investigate how your models access and potentially privilege knowledge from specific languages. Simultaneously, design and integrate decision-support mechanisms that help users critically evaluate imperfect AI outputs, especially in cross-cultural settings. This approach will foster more trustworthy and effective AI systems for diverse global communities.

Key insights

Trustworthy AI-mediated communication requires addressing both model biases and user reliance in cross-cultural contexts.

Principles

• NLP systems exhibit a socio-technical gap in practice.
• Cross-cultural interaction stresses AI trustworthiness.
• Trustworthy AI requires model and user-side interventions.

In practice

• Investigate language privileging in multilingual models.
• Develop decision-support tools for AI users.
• Evaluate user reliance on imperfect AI outputs.

Topics

• Trustworthy AI
• Multilingual NLP
• Cross-Cultural Communication
• AI-Mediated Communication
• Decision Support Systems
• Socio-Technical Gap

Best for: Research Scientist, AI Scientist, NLP Engineer, AI Ethicist

Related on AIssential

• The Multilingual AI Gap Is Not Closing. It Is Being Rebranded. — Linguistic inclusion in AI requires community control and governance, not just more data or better benchmark scores.
• Language Models Exhibit Inconsistent Biases Towards Algorithmic Agents and Human Experts — LLMs exhibit inconsistent biases, preferring human experts in stated trust but algorithms in revealed preference tasks.
• Evaluating Commercial AI Chatbots as News Intermediaries — Commercial AI chatbots exhibit high news accuracy but are undermined by regional language bias, retrieval dependency, and vulnerability to false premises.
• Chatbots Are Scheming 5x More — AI literacy is crucial for responsible creation and critical use of AI tools, emphasizing human choice and data integrity.
• Extending the Contact Hypothesis: Cross-Linguistic Evaluation of Religion and Nationality Bias When Prompting LLMs in German and Icelandic — LLMs exhibit religion and nationality biases, requiring cross-linguistic evaluation for fairness.
• The Hidden Risk of Polished AI Language — AI's linguistic closure can make weak evidence sound authoritative, posing a hidden risk beyond obvious hallucinations.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Paper Index on ACL Anthology.