资讯探索

认识一下 GPT-Red：OpenAI 为了使其模型更安全而打造的 LLM 超级黑客

原文：Meet GPT-Red: an LLM super-hacker OpenAI built to make its models safer

MIT Technology Review · score 43.8 · 7/15/2026, 10:09:37 AM

摘要

OpenAI has built an LLM super-hacker called GPT-Red that it uses as a sparring partner to help its other models boost their defenses against cyberattacks. Last week the company released the latest version of its flagship LLM, GPT-5.6. OpenAI says that training it against GPT-Red made the model its most robust release yet. GPT-Red automates…

原文摘要：OpenAI has built an LLM super-hacker called GPT-Red that it uses as a sparring partner to help its other models boost their defenses against cyberattacks. Last week the company released the latest version of its flagship LLM, GPT-5.6. OpenAI says that training it against GPT-Red made the model its most robust release yet. GPT-Red automates…

原始内容

OpenAI has built an LLM super-hacker called GPT-Red that it uses as a sparring partner to help its other models boost their defenses against cyberattacks. Last week the company released the latest version of its flagship LLM, GPT-5.6. OpenAI says that training it against GPT-Red made the model its most robust release yet. GPT-Red automates&#8230;