PeopleofSuperintelligence

Researcher

Jiseon Kim

Jiseon Kim is a natural language processing researcher who studies how well language models align with the plurality of human and societal values. She earned her doctorate at KAIST, advised by Alice Oh, and coauthored the Korean bias benchmark KoBBQ, published in TACL in 2024. Her papers on legislative bill similarity in 2021 and on lobbying positions in 2025 apply language models to political data, and she received a KAIST Graduate Student Outstanding Paper Award in 2024 and an MIT MISTI Global Seed Fund in 2025. In 2026, her Findings of ACL paper compared how language models judge moral rightness, predict human behavior and make their own decisions in relational moral dilemmas. She is a Distinguished Postdoctoral Fellow at the Vector Institute for Artificial Intelligence.

Timeline

  1. 2021

    Bill similarity at EMNLP

    She is first author of 'Learning Bill Similarity with Annotated and Augmented Corpora of Bills' at EMNLP 2021, written with Elden Griggs, In Song Kim and Alice Oh. The paper builds a human-labeled dataset of 4,721 bill-to-bill relationships.

  2. 2024

    The KoBBQ bias benchmark

    As a Ph.D. student in KAIST's School of Computing, advised by Alice Oh, she coauthors KoBBQ, a Korean bias benchmark for question answering, with Jiho Jin, Nayeon Lee, Haneul Yoo, her advisor and Hwaran Lee of NAVER AI Lab. The benchmark, published in TACL in 2024 and presented at ACL 2024, has 268 templates and 76,048 samples across 12 categories of social bias.

  3. 2024

    A KAIST outstanding paper award

    She receives a KAIST Graduate Student Outstanding Paper Award in 2024.

  4. 2025

    Lobbying positions across Congress

    She is first author of 'Measuring Interest Group Positions on Legislation: An AI-Driven Analysis of Lobbying Reports' in 2025, written with Dongkwan Kim, Joohye Jeong, Alice Oh and In Song Kim. Large language models and graph neural networks extract positions from lobbying activities, producing a dataset of 42k bills annotated with 279k bill positions of 12k special interest groups.

  5. 2025

    A MISTI Global Seed Fund

    She receives an MIT MISTI Global Seed Fund in 2025. Her research positions include MIT, the Max Planck Institute for Security and Privacy and NAVER AI Lab.

  6. May 2026

    Organizing the SocialLLM workshop

    In May 2026, she co-organizes the SocialLLM workshop on large language models for social reasoning and simulation at ICWSM 2026 in Los Angeles, with Xiangjue Dong, EunJeong Hwang and Alice Oh.

  7. July 2026

    Relational moral dilemmas at ACL

    In July 2026, her paper 'Machine Behavior in Relational Moral Dilemmas: Moral Rightness, Predicted Human Behavior, and Model Decisions' appears in Findings of ACL 2026 in San Diego, written with Jea Kwon, Luiz Felipe Vecchietti, Wenchao Dong, Jaehong Kim and Meeyoung Cha. Using the Whistleblower's Dilemma, it varies crime severity and relational closeness to compare how language models judge moral rightness, predict human behavior and make their own decisions.

  8. As of October 2026

    Postdoctoral fellow at Vector

    She is a Distinguished Postdoctoral Fellow at the Vector Institute for Artificial Intelligence, working with Tim G. J. Rudner. She is also an organizer of the SocialAgent workshop on large language models for social reasoning and simulation at NeurIPS 2026 in Atlanta.

Career

  1. NowVector InstituteDistinguished Postdoctoral Fellow

Education

  1. KAISTPhD

From the biography.

  1. Of 943 people shaping AI who worked at Google, 446 work there nowIncludes Vector Institute. Our estimates of where the 943 people shaping AI who worked at Google and Google DeepMind work now: 446 at Google and 452 across 278 other organizations.
  2. Nonprofit roles are 2.3 times as common in AI safetyIncludes Vector Institute. We estimate where people focused on safety, alignment and evaluation work. Nonprofit roles appear for 25.1% of safety people with known affiliations, compared with 11.0% in other areas.
  3. Universities (31%) and startups and other companies (25%) employ the largest shares of 2020–2026 entrantsIncludes Vector Institute. Among 819 people whose first recorded role started in 2020–2026 and who have a known current organization, 60% work in commercial sectors (25% at startups and other companies, 23% at frontier labs and 13% at big tech, with 9 in two), and 31% work at universities, compared with 66% for earlier entrants.
  4. The United States employs an estimated 48% of people shaping AIIncludes Vector Institute. We estimate where the people shaping AI work today, how universities, companies and research institutes split across countries, and how country shares have changed since the 1960s.

Sources and credits

Photo: We have no published photo of Jiseon Kim yet. Send us a free-licensed photo.

  1. Learning Bill Similarity with Annotated and Augmented Corpora of Bills - ACL Anthology aclanthology.org. Checked 21 October 2026.
  2. KoBBQ: Korean Bias Benchmark for Question Answering | Transactions of the Association for Computational Linguistics transacl.org. Checked 21 October 2026.
  3. Jiseon Kim - Vector Institute for Artificial Intelligence vectorinstitute.ai. Checked 21 October 2026.
  4. [2504.15333] Measuring Interest Group Positions on Legislation: An AI-Driven Analysis of Lobbying Reports arxiv.org. Checked 21 October 2026.
  5. Social Agents Workshop Series | NeurIPS, ACL & ICWSM 2026 social-llm-workshop.github.io. Checked 21 October 2026.
  6. Machine Behavior in Relational Moral Dilemmas: Moral Rightness, Predicted Human Behavior, and Model Decisions - ACL Anthology aclanthology.org. Checked 21 October 2026.