Maartje De Meulder

Maartje De Meulder, Davy Van Landuyt, Rehana Omardeen

In the era of AI-driven language technologies, the participation of deaf communities in sign language technology development, often framed as co-creation, is increasingly emphasized. We present a reflexive case study of two Horizon 2020 projects on sign language machine translation (2021- 2023), conducted with a EUD, a European-level deaf-led NGO. Using participant observation, internal documentation, and collaborative analysis among the authors, we interrogate co-creation as both a practice and a discourse. We offer five lessons for making co-creation consequential: 1) recognise and resource deaf partners invisible labor, 2) manage expectations via accessible science communication, 3) crip co-creation by dismantling structural ableism, 4) diversify participatory methods to address co-creation fatigue and intersectionality, and 5) redistribute power through deaf leadership. We contribute an empirically grounded account of how co-creation plays out in multi-partner AI projects, and actionable implications for design that extend to participatory AI with minoritized language and disability communities.

Aashaka Desai, Maartje De Meulder, Julie A. Hochgesang et al.

Growing research in sign language recognition, generation, and translation AI has been accompanied by calls for ethical development of such technologies. While these works are crucial to helping individual researchers do better, there is a notable lack of discussion of systemic biases or analysis of rhetoric that shape the research questions and methods in the field, especially as it remains dominated by hearing non-signing researchers. Therefore, we conduct a systematic review of 101 recent papers in sign language AI. Our analysis identifies significant biases in the current state of sign language AI research, including an overfocus on addressing perceived communication barriers, a lack of use of representative datasets, use of annotations lacking linguistic foundations, and development of methods that build on flawed models. We take the position that the field lacks meaningful input from Deaf stakeholders, and is instead driven by what decisions are the most convenient or perceived as important to hearing researchers. We end with a call to action: the field must make space for Deaf researchers to lead the conversation in sign language AI.