When the AI Dam Breaks: From Surveillance to Game Theory in AI Alignment
"审慎对齐" 相关文章
When the Dam Breaks: From Surveillance to Game Theory in AI Alignment
When the Dam Breaks: Could Game Theory Create An Alternative Alignment Approach?
故意“装菜”答错问题，AI已能识别自己“正在被测试”
The Sweet Lesson: AI Safety Should Scale With Compute
On Deliberative Alignment
OpenAI trained o1 and o3 to ‘think’ about its safety policy
