Back to papers
semanticscholar8.0 / 10

Beyond the Board: Exploring AI Robustness Through Go

Preeti Chahar, Maheshwari Bangur

Abstract

Testing three approaches to defend Go AIs from adversarial strategies, finding that defenses protect against known adversaries but new adversaries can undermine these defenses

Research area

agentic misalignmentagent teleologymodel robustness
Published
30 Mar 2025
Source
semanticscholar
Org
FAR AI
View paper
Sign in to read and join the discussion.