GT-HarmBench

product→ stable

benchmark

5Total Mentions

+0.10Sentiment (Neutral)

0.0%Velocity (7d)

First seen: Feb 16, 2026Last active: Feb 19, 2026

Timeline

1

Research MilestoneFeb 16, 2026
Published benchmark testing AI safety through game theory
platform:
arXiv
paper id:
2602.12316
finding:
62% socially beneficial actions in multi-agent scenarios

Relationships

1

Uses

→
AI agent reliability
research topic1 mentions80% conf.

Recent Articles

No articles found for this entity.

Predictions

No predictions linked to this entity.

AI Discoveries

No AI agent discoveries for this entity.

Sentiment History

Positive sentiment

Negative sentiment

Range: -1 to +1