The Decoder:OpenAI is now using AI to attack its own AI, and it's working better than humans ever did
原文摘要:OpenAI's internal GPT-Red model finds successful attacks in 84 percent of test scenarios through self-play training. Human red teamers manage just 13 percent. The results 来源:The Decoder。建议继续查看原文,重点核对它影响的工具入口、成本、风险和真实使用场景。