Unverified50% confidenceReleaseExact time

OpenAI披露了一个名为GPT-Red的内部对抗性模型，专门用来攻击AI智能体的红队安全工具

1

Sources

50%

Confidence

Medium-term (~90 days)

Relevance

7/16/2026

First Seen

Valid until: 10/14/2026

Sources

GPT-Red：OpenAI用AI攻击自家模型的红队安全工具

Related Claims

UnverifiedOpenAI将GPT-Red与已部署模型完全隔离，以避免攻击能力外泄或被开放给潜在对手81% similar VerifiedOpenAI发布了GPT-5.5 Cyber，这是一款专门面向关键基础设施防御人员的AI模型75% similar UnverifiedOpenAI的InstructGPT、Anthropic的Constitutional AI以及Meta的Llama系列都采用了类似的安全对齐流程72% similar

Cite This Claim

Stable URI

https://kongchang.com/claim/533953

API

curl https://kongchang.com/api/v1/knowledge/claims/533953

MCP

get_claim(id=533953)