Position Paper Accepted at ICML 2026
Excited to share that our position paper “Breaking the Dual Curse of Multilingual AI Requires Socio-Technical Guardrails, Not Post-Hoc Alignment” has been accepted to the 43rd International Conference on Machine Learning (ICML 2026) — Position Paper Track.
We argue that multilingual AI safety cannot be retrofitted through post-hoc alignment. Our systematic review documents a dual curse:
- Safety collapse in low-resource languages: harmful content generation rises to 35% in low-resource languages, versus 1% in English.
- Instruction-following collapse: capability drops sharply across the same languages.
- Broken safety pipelines: across 207 studies, reward models achieve only 49–50% accuracy in low-resource languages — equivalent to random chance — undermining RLHF and downstream safety filtering.
Rather than another patch on top of a broken pipeline, we call for socio-technical guardrails built in from pre-training: community-led harm specification, multilingual evaluation metrics that jointly balance security and usability, and interventions that treat multilingual safety as a first-class design constraint.
Grateful to my co-authors — Pureheart Ogheneogaga Irikefe, Adaku Uchendu, Umniya Najaer, Cornelius Adejoro, Patrice Sterling, and my advisor Dr. Dongwon Lee — for a genuinely transdisciplinary collaboration.
Resources:
I completed my Ph.D. in Informatics at Penn State University (defended May 2026; formal conferral August 2026), where I conducted research at the PIKE Research Lab under Dr. Dongwon Lee and the College of IST. Starting August 2026, I will join the Department of Information Science at the College of Media, Communication and Information (CMDI), University of Colorado Boulder, as a Tenure-Track Assistant Professor and founding Director of the Secure and Ethical AI Lab (SEAL). My research advances trustworthy and equitable AI for the world’s languages and communities — spanning multilingual NLP, low-resource and dialectal language technology, AI safety, and information integrity, with work extending across 70+ languages. I have authored 14+ peer-reviewed papers with 315+ citations in premier venues including ACL, EMNLP, NAACL, ICML, KDD, and IEEE.
My doctoral research focuses on bridging the digital language divide through transfer learning, classification (NLU), generation (NLG), adversarial attacks, and developing end-to-end AI pipelines using RAG and Agentic AI workflows for combating multilingual threats. Drawing from my Grenadian background and knowledge of local Creole languages, I bring a global perspective to AI challenges, working to democratize state-of-the-art AI capabilities for underserved linguistic communities worldwide. My mission is to develop robust multilingual multimodal systems and mitigate evolving security vulnerabilities while enhancing access to human language technology through cutting-edge solutions.
As an NSF LinDiv Fellow, I conduct transdisciplinary research advancing human-AI language interaction for social good. I actively mentor 5+ research interns and teach Applied Generative AI courses. Through industry experience at Lawrence Livermore National Lab, Interaction LLC, and Coalfire, I bridge academic research with practical applications in combating evolving security threats and enhancing global AI accessibility. I see multilingual advances and interdisciplinary collaboration as a competitive advantage, not a communication challenge. Beyond research, I stay active through dance, fitness, martial arts, and community service.