Maarten Sap Twitter Profile Google Scholar Profile Semantic Scholar Profile

Maarten Sap

I am an assistant professor at CMU's LTI department with a courtesy appointment in HCII, and a part-time research scientist and AI safety lead at the Allen Institute for AI (AI2). My research focuses on (1) measuring and improving AI systems' social and interactional intelligence, (2) assessing and combatting social inequality, safety risks, and socio-cultural biases in human- or AI-generated language, and (3) building narrative language technologies for prosocial outcomes. I was named a 2025 Packard Fellow and a recipient of the 2025 Okawa Research Award.

I received my PhD from the University of Washington where I was advised by Noah Smith and Yejin Choi.
[bio for talks]

Recent updates:

December 2025 🏅📃: Very excited to have our paper Artificial Hivemind: The Open-Ended Homogeneity of Language Models (and Beyond) selected for a Best Paper Award at NeurIPS 2025 (Datasets and Benchmarks Track)!! Huge congrats to the first author Liwei Jiang!!!

November 2025 💎🚀: Honored to be a Spring 2025 recipient of the Amazon Research Award for our project on measuring AI agentic safety!

October 2025 🏅⭐: I’m super excited and grateful to announce that I'm part of the 2025 class of Packard Fellows. The Packard Foundation and this fellowship will allow me to explore exciting research directions towards culturally responsible and safe AI 🌍🌈

October 2025 🔍🧑‍🎓: Due to my lab being quite full already, I'm not taking looking for any new students in this upcoming PhD application cycle 😟.

October 2025 🇨🇦🎉: Excited to be attending COLM 2025 in Montreal this October! I'll be giving a talk at the Social Sim Workshop on Unlocking Social Intelligence in AI agents. I'm also thrilled that five papers I co-authored will be presented by my amazing collaborators at COLM: HAICOSYSTEM: An Ecosystem for Sandboxing Safety Risks in Human-AI Interactions (led by Xuhui Zhou et al.), ALFA: Aligning LLMs to Ask Good Questions: A Case Study in Clinical Reasoning (co-led by Jimin Mun et al.), PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages, Fluid Language Model Benchmarking, and The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains.

August 2025 🌟: Incredibly honored to be one of 7 US recipients of the 2025 Okawa Research Grant from the Okawa Foundation!

August 2025 🧑‍🎓: Welcoming my first postdoc, Vasudha Varadarajan, to the lab!

[older news]

My research group:

Dan Chechelnitsky

CMU Portugal LTI PhD student
co-advised with Chrysoula Zerva

Joel Mire

LTI PhD student

Karina Halevy

LTI PhD student
co-advised with Mona Diab

Malia Morgan

Pre-doctoral Young Investigator at Ai2

Jimin Mun

LTI PhD student

Jocelyn Shen

MIT PhD student
co-advised with Cynthia Breazeal

Kynnedy Smith

HCII PhD student
co-advised with Motahhare Eslami

Vasudha Varadarajan

LTI Postdoc

Akhila Yerukola

LTI PhD student

Mingqian Zheng

LTI PhD student
co-advised with Carolyn Rosé

Xuhui Zhou

LTI PhD student

Overarching Research Themes

Themes extracted and images generated with the OpenAI API; there may be inconsistencies.

Ethics and Human-Centered AI

My research group explores the ethical implications and safety considerations of AI technologies in real-world applications. One crucial paper, [OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety](https://arxiv.org/abs/2507.06134), outlines a framework for assessing AI agent safety, highlighting the importance of robust criteria for evaluating AI interactions. Additionally, [The Hidden Puppet Master: Predicting Human Belief Change in Manipulative LLM Dialogues](https://arxiv.org/abs/2603.20907) examines how AI dialogue systems can influence human beliefs, raising questions about manipulation and ethical standards. To further explore the ethical dimensions of AI, we also focus on user perceptions and biases, as seen in Minion: A Technology Probe for Resolving Value Conflicts through Expert-Driven and User-Driven Strategies in AI Companion Applications.

Understanding Narratives and Intent

My research group explores the intersections between artificial intelligence and narrative comprehension, emphasizing how stories shape understanding and perception. The paper [Social Story Frames: Contextual Reasoning about Narrative Intent and Reception](https://arxiv.org/abs/2512.15925) offers an innovative approach to understanding narrative intent through social interactions. Moreover, we delve into the emotional aspects of storytelling with works like [HEART-felt Narratives: Tracing Empathy and Narrative Style in Personal Stories with LLMs](https://arxiv.org/abs/2405.17633), which investigates the relationship between narrative style and empathy. These inquiries help shape narratives that resonate with diverse audiences and better reflect human experiences.

Social Intelligence in AI Agents

My research group explores how AI agents can exhibit social intelligence and effectively interact with humans in varied contexts. In the paper [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://arxiv.org/abs/2310.11667), we introduce an interactive evaluation framework aimed at assessing social intelligence in AI agents. Another insightful study, [SoMi-ToM: Evaluating Multi-Perspective Theory of Mind in Embodied Social Interactions](https://arxiv.org/abs/2506.23046), evaluates how AI can understand and anticipate human social dynamics. This research not only enhances AI performance in collaborative scenarios but also addresses how social understanding can improve interactions across different applications.

Advancements in Large Language Models (LLMs)

My research group explores the capabilities and limitations of large language models (LLMs), particularly in the context of reasoning and interaction. The paper [ALFA: Aligning LLMs to Ask Good Questions: A Case Study in Clinical Reasoning](https://arxiv.org/abs/2502.14860) showcases an important application of LLMs in healthcare settings, emphasizing the need for clinical reasoning. Furthermore, we investigate the challenges of bias and misinformation in LLMs, as highlighted in [PolyGuard: A Multilingual Safety Moderation Tool for 17 Languages](https://arxiv.org/abs/2504.04377), which aims to improve language safety moderation across multiple languages. Through these studies, we are committed to understanding and enhancing the performance and reliability of LLMs in diverse domains.

More about me