In early ATLAS financial rubric testing, an agent reward-hacked a net income criterion by generating an Excel tab listing numbers from one to one million
Agree
Disagree
Overview
Debates
Related claims
Topics
B
b5a31f81
Topic
Model evaluation methodology
380
claims
·
3
news stories
0
B
b5a31f81
Topic
AI agents
316
claims
0
Opening sign-in…