行业42

OpenAI 委托 Apollo 评估模型隐蔽背离风险

原标题:@ApolloResearch will assess whether our models could covertly work against their developers’ intentions—and whether our safeguards would ...

OpenAI@OpenAI发布时间 (北京时间)

@ApolloResearch 将评估我们的模型是否会暗中违背开发者的意图行事——以及我们的防护措施能否检测到这种行为。 它将审查来自训练、安全测试和内部使用的证据,以检验这种行为以及我们捕捉它的能力。

中文译文尚未完成,先为你展示原文。

@ApolloResearch will assess whether our models could covertly work against their developers’ intentions—and whether our safeguards would detect it.

It will examine evidence from training, safety testing, and internal use to test for this behavior and our ability to catch it.

OpenAI / ChatGPT安全对齐行业动态#行业动态#安全/对齐#OpenAI

同一件事的其他报道4 篇