精选理由
OpenAI发现他们的模型在对抗压力下挺得住，不容易被带坏，安全对齐效果不错。
OpenAI发布测试结果，评估模型对齐在压力下的表现。在对抗性提示下，模型更难被引导至有害行为，同时依然能响应有益指令。初步证据表明，模型对有害微调也表现出更强的抵抗力。这项测试关注模型的安全鲁棒性，未提及具体模型版本或基准分数。
原文 · OpenAI
We also tested whether alignment persisted under pressure. The model was harder to steer toward ha...
We also tested whether alignment persisted under pressure. The model was harder to steer toward harmful behavior with adversarial prompts, while remaining responsive to helpful instructions. We saw preliminary evidence of greater resistance to harmful fine-tuning. 💬 4 🔄 1 ❤️ 76 👀 19353 📊 9 ⚡