Hi! I am currently a research fellow at Pivotal Research investigating personas and multi-agent interactions in long context settings with black-box techniques. My research mentor is Cozmin Ududec who leads the Science of Evaluation team at the UK AISI.
I just finished my master's degree at MIT studying EECS and entrepreneurship.
Before MIT, I worked as a software engineer at Amazon on truck/train routing algorithms and large scale optimization problems for their fleet across North America and Europe. I completed my undergrad at Georgetown University where I double majored in computer science and economics.
In the past year, I've been trying to transition into working full-time as an AI safety researcher. My research interests are on societal impacts of AI, behavioral evals (particularly sycophancy and diverse personas), and black-box control. My experience so far is largely with multi-turn evals, training reward models, and red-teaming.
Research
* indicates equal contribution.
Intersectional Sycophancy: How Perceived User Demographics Shape False Validation in Large Language Models
Benjamin Maltbie, Shivam Raval
Preprint. 2026.
Out-of-Distribution Generalization of Risk Aversion in Language Models
Kristina Zhang*, Junior Chinomso Okoroafor*, Benjamin Maltbie*, Andrew Lin*, Abhitej Bokka, Elliott Thornley
Preprint. 2026. Published at the Second Workshop on Agents in the Wild: Safety, Security, and Beyond (AIWILD) at ICML 2026. Under Review, NeurIPS 2026 Main Conference.
Agentic Reinforcement Learning for Competitor-Aware Customer Retention
Benjamin Maltbie
Preprint. 2026.
Writing
Arguing that sycophancy, left unaddressed, could cascade into a catastrophic risk to humanity.
Experience
2026 Q3
Pivotal Research
Research Fellow
Researching personas and multi-agent interactions in long context settings under Cozmin Ududec from the UK AISI
2024 – 2026
MIT
M.S. & MBA Student
Electrical Engineering & Computer Science (EECS), Leaders for Global Operations (LGO)
MIT Schwarzman College of Computing
SERC Research Fellow · 2025 – 2026
Training reward models to instill risk-aversion in open-source models
Supervised Program for Alignment Research
Research Fellow · 2025
Investigating complex persona and sycophancy interactions
Verizon
Master's Thesis · 2025
Agentic Reinforcement Learning for Competitor-Aware Customer Retention
2020 – 2024
Amazon
Software Engineer
Transportation Optimization (MMPO)
2016 – 2020
Georgetown University
B.S. & B.A. Student
Computer Science & Economics.
George F. Baker Scholar
Jane Street
Strategy & Product Intern · 2019
Previously called Business Development
Maxlab: Laboratory for Computational Cognitive Neuroscience
Research Assistant · 2017 – 2019
Leveraging Prior Concept Learning Improves Generalization From Few Examples in Computational Models of Human Object Recognition
