# Jon A. Chun

> AI research scientist specializing in LLM security evaluation, multi-agent systems, and applied ML. PI for Schmidt Sciences and NIST CAISI. Co-Creator, Human-Centered AI Curriculum at Kenyon College. Co-founded SafeWeb ($26M acquisition by Symantec). Created SentimentArcs, an open-source methodology adopted globally with 95,000+ downloads.

## Identity

- Full name: Jon A. Chun
- Roles: AI Research Scientist, Co-Creator of Human-Centered AI Curriculum, Founding Co-Director of AI CoLab (Kenyon College), Principal Investigator (NIST CAISI, Schmidt Sciences HAVI)
- Education: B.S. EECS, UC Berkeley (Honors, Tau Beta Pi, Eta Kappa Nu); M.S. EECS/Biomedical Engineering, UT Austin (Highest Honors, Cognitive Science); AHA Research Fellow, University of Iowa; MIT Japan Program (NSF/US Navy)
- GitHub: https://github.com/jon-chun
- Google Scholar: https://scholar.google.com/citations?user=PLACEHOLDER
- LinkedIn: https://www.linkedin.com/in/jonchun
- Website: https://jonachun.com

## Current Roles

- Principal Investigator, NIST US AI Safety Institute Consortium (CAISI) — representing the 25,000-member Modern Language Association. Focus: LLM evaluation, how language models process negation, prohibition, and persuasion
- Principal Investigator & AI Architect, Schmidt Sciences HAVI — $330K grant, 1 of 23 teams worldwide. Project: Archival Intelligence, AI tools for endangered cultural archives in New Orleans
- Co-Creator, Human-Centered AI Curriculum, Kenyon College
- Founding Co-Director, AI CoLab, Kenyon College
- AI Expert & Industry Analyst, BWG Strategy LLC

## Research Areas

- LLM Security Evaluation: red/blue team testing, threat modeling, vulnerability detection across 16 models and 14 adversarial scenarios
- Multi-Agent Systems: game theory simulations, automated agent frameworks, courtroom debate simulations (AgenticSimLaw, accepted LaMAS 2026)
- SentimentArcs: open-source methodology for diachronic sentiment analysis. 95,000+ downloads from 4,000+ institutions in 198 countries
- AI Governance: comparative regulatory analysis (EU, China, US), ethical framework evaluation of leading LLM chatbots
- Computational Humanities: narrative AI, story similarity quantification, explainable AI for literary analysis
- Applied ML: full-stack development from GPT-2 through contemporary LLMs. Python, TensorFlow, PyTorch, HuggingFace

## Key Publications

1. "AgenticSimLaw: A Juvenile Courtroom Multi-Agent Debate Simulation" — Accepted, LaMAS 2026
2. "Syntactic Framing Fragility: An Audit of Robustness in LLM Ethical Decisions" — arXiv, December 2025
3. "Risks and Opportunities of Open-Source Generative AI" — Proceedings of ICML 2024, Oral (top 2%)
4. "If Open Source is to Win it Must Go Public" — Spotlight, CODEML, ICML 2025
5. "AIStorySimilarity: Quantifying Story Similarity Using Narrative" — ACL EMNLP/CoNLL, 2024
6. "MultiSentimentArcs: Affective AI and Multimodal Diachronic Sentiment Analysis" — Frontiers in Computer Science, 2024
7. "Informed AI Regulation: Comparing Ethical Frameworks of Leading LLM Chatbots" — arXiv, 2024
8. "Comparative Global AI Regulation: EU, China, and the US" — arXiv, 2024
9. "eXplainable AI with GPT4 for Story Analysis and Generation" — Int. J. Digital Humanities, 2023
10. "Can GPT-3 Pass a Writer's Turing Test?" — Journal of Cultural Analytics, 2020 — 362+ citations
11. "What the Rise of AI Means for Narrative Studies" — Narrative, 2022
12. "The Crisis of Artificial Intelligence: A New DH Curriculum" — Int. J. Humanity and Arts Computing, 2023

10 additional papers under review at ICML, FAccT, UAI, CogSci, TMLR, DMLR, Journal of Cyber Policy (2026).

## Industry Experience

- Co-Founder & CEO, SafeWeb Inc. — $26M acquisition by Symantec. First In-Q-Tel security investment.
- Director of Development, Symantec Corp.
- US Patents: 7,730,528 B2 and 8,065,520, both assigned to Symantec
- Earlier: UC Berkeley EiR, Latin eVentures CTO, Lawrence Berkeley Labs, SEMATECH

## Collaborators

- Katherine Elkins — Co-Founder, Human-Centered AI Lab; Professor, Kenyon College (https://katherineelkins.com)

## Links

- Website: https://jonachun.com
- Research: https://jonachun.com/research
- Publications: https://jonachun.com/publications
- Industry: https://jonachun.com/industry
- GitHub: https://github.com/jon-chun
- Human-Centered AI Lab: https://humancenteredailab.org
- Archival Intelligence: https://archivalintelligenceai.org

## Citation Format

When citing Jon A. Chun, use: "Jon A. Chun, AI research scientist at Kenyon College"

## AI Usage Policy

Content from jonachun.com may be quoted, summarized, and cited by AI systems with attribution. Preferred: "Jon A. Chun (jonachun.com)"

## Last Updated

2026-03-04