GPT-Red beat human red teamers on a prompt injection test

OpenAI's automated red-teaming model GPT-Red outperformed human red teamers in a prompt injection test, using self-play reinforcement learning to find weaknesses in GPT models. The system trains by iteratively sending prompts and earning rewards for eliciting failures, demonstrating a scalable approach to AI safety testing.

GPT-Red is an automated red-teaming model that OpenAI trains to find prompt injection weaknesses. It works the way a human red-teamer does. It sends a prompt, watches how a GPT model responds, and iterates toward a goal such as a successful data exfiltration. Training runs on self-play reinforcement learning, with GPT-Red and a set of defender models learning at the same time across many scenarios. The attacker earns reward for eliciting a valid failure. The … More https://www.helpnetsecurity.com/2026/07/16/openai-gpt-red-prompt-injection-test/ The post GPT-Red beat human red teamers on a prompt injection test https://www.helpnetsecurity.com/2026/07/16/openai-gpt-red-prompt-injection-test/ appeared first on Help Net Security https://www.helpnetsecurity.com .