Paper 005·Red Teaming·January 2026

Large Empirical Case Study: Go-Explore adapted for AI Red Team Testing

M. Bhatt, A. Wood, I. Habler, A. Al-Kahfah

Abstract

Adapts Go-Explore to agent red teaming and shows that random-seed variance can dominate algorithm choices, making multi-seed evaluation essential for defensible comparisons.

Research record

This page summarizes the public research record and links to the authoritative paper source. Propose a replication or report a discrepancy to b1oo@shrewdsecurity.io.