This really nicely echoes a response to Eliezer Yudkowsky that I saw: if you always just play the odds on stuff like this, you're gonna be wrong about the real

This really nicely echoes a response to Eliezer Yudkowsky that I saw: if you always just play the odds on stuff like this, you're gonna be wrong about the real breakthroughs, which by definition defy the odds. There are plenty of reasons to be skeptical, but I remain hopeful. All the theoretical critiques are extra interesting if this is real and the data is basically accurate.