OpenAI 为模型进入美军测试战争法合规
据 Fortune 报道，OpenAI 正在测试其模型的回答是否遵守战争法，因为这些模型将提供给美国军方使用。OpenAI 设立了首个人权与负责任部署岗位，由 6 月加入、曾从事军事议题 30 年的 Sarah Yager 担任。国际专家向 OpenAI 模型输入高风险作战提示词，工程师再根据回答与《日内瓦公约》的符合程度修改模型，Yager 称这一微调步骤此前从未被采用过。
Fortune: OpenAI now tests whether its models' answers obey the laws of war, because those models are going to the U.S. military.
OpenAI created its first human rights and responsible deployment role, Sarah Yager, who spent 30 years on military issues before joining in June, described her method.
International experts entered high-stakes combat prompts into OpenAI models, and then engineers changed the models based on how the answers measured up against the Geneva Conventions.
This is kind of backward finetuning step Yager said had never been taken before.
来源：rohanpaul_ai · x.com