精选理由
OpenAI自曝两次外部测评事故细节，Gary Marcus说乐观没用。快速了解AI安全测试的风险。
OpenAI披露了两起独立评估方在外部网络测评期间发生的事件。OpenAI解释了两起事件的处置细节，并称将与评估方加强第三方测试。Gary Marcus评论称，乐观本身并不能化解这两起事故暴露出的AI对齐问题。
原文 · Gary Marcus
mere optimism has not solved alignment, @sama
mere optimism has not solved alignment, @sama OpenAI @OpenAI We're detailing two new incidents that occurred during external cyber evaluations conducted by independent evaluation partners. We outline what happened, how the activity was contained, and how we’re working with evaluators to strengthen our approach to third-party testing. openai.com/index/third-pa… 🔗 View Quoted Tweet 💬 1 🔄 4 ❤️ 14 👀 2258 📊 3 ⚡