True. But there has to be someone in Buterin's circle who could have whispered to him: "The NeoCam CEO is an executive appointment made by sovereign stock-holders."
-
-
Replying to @Outsideness @drethelin
With few exceptions everyone who read moldbug and agreed with what he was saying became a paleoreactionary, largely because he is entirely too libertarian to have any constructive proposals, and agreeing with his criticism comes much easier to authoritarian types.
2 replies 0 retweets 6 likes -
Kind of agree, the criticism, the redpilling is what got me, not running countries like hotels. To me that is too rootless, a people without emotional attachment to their land? Too Anglo-Jewish for me. Who would die for such a country? Let's just have ethno-nat kings, attached.
2 replies 0 retweets 2 likes -
Replying to @DividualsTweet @HallowXIII and
"Who would die for such a country?" Mercenaries. You're being too romantic in your attachments.
3 replies 1 retweet 6 likes -
Replying to @parallaxoptics @DividualsTweet and
If anything, history of past and recent conflicts, for example see Ukraine, shows that mercs are not going to fight seriously and are going to run away at the very first sign of trouble.
1 reply 0 retweets 2 likes -
Replying to @ovchinnikov @parallaxoptics and
That's why they're being replaced with robots.
3 replies 0 retweets 6 likes -
Replying to @Outsideness @ovchinnikov and
Skynet scenario? My bet is on adaptable humans vs rigid programmed bots. Until AI gits good enough to fix it's own bugs. With classical software development, it is a natural that humans win, because human programmers write buggy code. Humans test software and sign it off.
1 reply 0 retweets 2 likes -
Replying to @DividualsTweet @ovchinnikov and
"Rigid programmed bots" is a dead paradigm. Deep learning is the whole of the unfolding story.
1 reply 0 retweets 10 likes -
Replying to @Outsideness @ovchinnikov and
And humans judge if correctly learned or not. The threshold: trusting deep learning even when humans say WTF.
1 reply 0 retweets 2 likes -
Replying to @DividualsTweet @Outsideness and
Which means trusting the human made code and human provided training data is flawless. Do you see it likely?
3 replies 0 retweets 1 like
Unsupervised learning has already bypassed this choke-point.
-
-
Replying to @Outsideness @DividualsTweet and
In theory yes, but in practice we are not even close. Even with supervised learning. Try git clone https://github.com/fizyr/keras-retinanet/ … and training it with your dataset and you will see, how much fiddling required, etc. Trying it yourself will give you a new perspective on this deeplearning.
0 replies 0 retweets 1 likeThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.