Skip to content

Matthew Kowal

FAR.AI

1 work in GUIDE

In GUIDE

2026 It's the Thought that Counts: Evaluating the Attempts of Frontier LLMs to Persuade on Harmful TopicsKowal et al. · arXiv

Researcher at FAR.AI working on large language model safety, alignment, and evaluation of persuasion capabilities.

Sources in this wiki

2026-kowal-persuade-harmful-llms

Topics

llm-safety, llm-alignment