My daily driver is Debian. It's practical, efficient, stable, and with just a few commands clipped out of blog articles, it morphs into whatever weird silly thing I happen to need it to be, this year.
Yeah. I've beaten Castle Wolfenstein. The secret turned out to be shooting all of the Nazis. Also a lot of secret doors, one in particular leading to a particularly big gun.
I want a world for my kids where our solution is nuanced, and education and growth oriented.
But if my dream isn't possible, I'll settle for a world where as many Nazis as possible are just dead.
I don't believe that you can use traditional algorithms to teach the car street driving, because there are to many different variations...
Even if your autopilot is 99% correct and you drive 20000km a year, you still drive wrong 200km of it.
Exactly!
And this is why, if the problem is solveable, it must be solved by learning models shepherded by expert engineers. The LLMs can take care of the long boring stretches, freeing skilled engineer time to fine-tune an LLM algorithm hybrid for the tricky bits.
I'm inclined to believe the problem is solveable, but since I'm not selling anything, I'm allowed to say "if". Heh.
Yeah. Current generation learning models can do impressive things in the hands of a skilled engineer, but Elon is leading a round of class warfare against skilled engineers right now.
Shareholders need to decide which they really want to bet on to win.
Sometimes I'm reminded that there's always a chance that they go submarine diving or some such with another overconfident crony who thinks their skills got them where they are today.
My brain did the same. Now I'm like "wel sh.....t!"