RT @DavidSacks: Is alignment safe? If you look at what Anthropic is actually doing, “alignment” does not mean training frontier models to…
Hesam(@Hesamation)87 字 (约 1 分钟)
50
文章讨论Anthropic的对齐方法存在信息缺失，未明确解释技术细节与实践影响。
入选理由：Anthropic的对齐训练不依赖传统指令遵循，而是通过Claude Constitution培养模型自我意识
FeaturedTweet#AI对齐#Anthropic#机器学习英文