Annabella Sakunkoo

HCI + NLP + AI safety researcher, Davidson Ambassador, pianist

Annabella Sakunkoo

Hey! I'm Annabella Sakunkoo, a researcher and student at Stanford OHS. I also take college coursework at Stanford, UCLA, and colleges such as Foothill and Santa Ana in computer science, statistics, linguistics, microeconomics, and fine arts, with a 4.00/4.00 GPA.

In 2026, I was selected as one of a selected few fellows in the Oxford AI Safety Core Fellowship at Oxford University — the youngest and only non-Oxford student admitted. In 2024 and 2025, I was selected for the MIT Beaver Works Summer Institute in Biotechnology and Cognitive Assistants (AI/NLP).

My research encompasses 2 main areas. I investigate fairness, bias, and safety in large language models. I'm also a builder. I design and create creative, educational, and joyful human/AI interaction systems for learning and wellbeing . My innovatived work has won a Tech Achievement Award from Stanford School of Medicine.

My research has been accepted for poster and demo presentations, talks, and panel discussion at ACM UIST in Korea, CHI in Japan and Spain, ALP CHI at ETH Zürich, ACL in Austria and California, and NAACL, where I also received an award. I'm a Yale Entrepreneurial Society Fellow and a Davidson Ambassador (one of only 20 students selected nationally).

Outside of research, I'm a pianist who has performed at Carnegie Hall (Zankel and Weill), Oxford University, Harvard University, Beethoven Haus in Germany, and the Opéra Bastille in Paris. I've taught low-income, marginalized students, tutored on Schoolhouse.world (where I was rated #18 most "Super Helpful" tutor), run the Stanford OHS World Languages and Philosophies Society, worked as a student leader of the Stanford iWiSH program at the Stanford Medical School , and spent some lovely time filming Disney commercials and Amazon Prime's Show.

news

Apr 2026 Our work, Through the Looking Glass of Multilingual AI, is accepted to ACL 2026 in San Diego!
Apr 2026 Our papers and posters, From Clicking to Moving: Embodied Micro-Movements as a New Modality for Data Literacy Learning and Through the Looking Glass of Multilingual AI, are presented at ACM CHI 2026 in Barcelona! I was also one of 5 panelists on Teaching and Practice in the CHI Data Literacy Workshop, organized by Northwestern and Stanford!
Feb 2026 Selected as one of 15 fellows for the Oxford AI Safety Core Fellowship at Oxford University — the youngest and only non-Oxford student admitted to the programme.
Jan 2026 Our work is accepted for demo presentation at ALP CHI 2026 at ETH Zürich, Switzerland.
Nov 2025 Our work SingLing: Learning Languages by Singing Code-Switched Lyrics is accepted to ACM UIST 2025!
Aug 2025 Our studies, Name of Thrones: How Do LLMs Rank Student Names in Status Hierarchies Based on Race and Gender? and Lost and Found: Computational Quality Assurance of Crowdsourced Knowledge on Morphological Defectivity in Wiktionary, are presented at ACL 2025 in Vienna!
Jul 2025 Delivered a 20-minute talk on Name of Thrones: How Do LLMs Rank Student Names in Status Hierarchies Based on Race and Gender? at International Conference on Computational Social Science (IC2S2) in Sweden.
Jun 2025 Won the Best Technical Achievement Award at Stanford Medicine.
May 2025 Received the NAACL D&I Award and presented research at the North American Chapter of the Association for Computational Linguistics.
Apr 2025 Selected for MIT Beaver Works Summer Institute in Cognitive Assistants (AI/NLP).
Apr 2025 Presented our work at CHI'25 in Yokohama, Japan. Also attended a tutorial on Interaction Techniques by Brad Myers
Apr 2025 Presented our work at New England NLP Symposium at Yale University
Mar 2025 Finalist in the North American Computational Linguistics Open Competition.
2024-2025 Selected as a Davidson Ambassador

selected publications

Through the Looking Glass of Multilingual AI ACL SRW
Through the Looking Glass of Multilingual AI: Contrasting Language- and Name Script-Dependent Ethnic Hierarchies in GPT and DeepSeek
In ACL 2026 SRW, ACL 2026, San Diego, USA.
From Clicking to Moving ACM IDC
From Clicking to Moving: Embodied Micro-Movements as a New Modality for Data Literacy Learning
In ACM IDC 2026, Brighton, UK.
Name of Thrones ACL 2025
Name of Thrones: How Do LLMs Rank Student Names in Status Hierarchies Based on Race and Gender?
In ACL 2025 Proceedings of the 20th Workshop on Innovative Use of NLP for Building Educational Applications (BEA), ACL 2025, Vienna, Austria.
Lost and Found ACL SRW
Lost and Found: Computational Quality Assurance of Crowdsourced Knowledge on Morphological Defectivity in Wiktionary
In ACL 2025 SRW, ACL 2025, Vienna, Austria.
SingLing UIST 2025
SingLing: Learning Languages by Singing Code-Switched Lyrics
In Adjunct Proceedings of the 38th Annual ACM Symposium on User Interface Software and Technology (UIST '25), Article 163, 1–3.
Lingdex RANLP 2025
Lingdex: Leveraging LLMs to Structure and Explore Linguistic Olympiad Puzzles for Learning and Teaching Linguistics
In Proceedings of the 15th International Conference on Recent Advances in Natural Language Processing - Natural Language Processing in the Generative AI Era.