GPT-Red

ai model→ stable

GPT-Red is an internal AI red-teaming model developed by OpenAI to autonomously attack its own systems, finding vulnerabilities in 84% of test scenarios compared to 13% for human testers.

1Total Mentions

+0.70Sentiment (Very Positive)

+1.2%Velocity (7d)

View subgraph

First seen: Jul 16, 2026Last active: 14h ago

Signal Radar

Five-axis snapshot of this entity's footprint

live

Loading radar…

Mentions × Lab Attention

Weekly mentions (solid) and average article relevance (dotted)

mentionsrelevance

Loading timeline…

Timeline

Research MilestoneJul 15, 2026
GPT-Red demonstrated 84% attack detection rate vs 13% for humans, detailed by MIT Technology Review
View source

Relationships

Developed

←
OpenAI
company1 source100% conf.

Uses

←
GPT-4o
ai model1 source90% conf.

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