I think most of it is money well spent, because we will eventually reap the rewards, and the rewards are significant. But it goes to show just how hard it is to build AI systems that interface with a wide range of situations in the real world with little margin for error.
-
-
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
It’s about 40Bn over the last 5/6 years (Crunchbase AV hub) but you could add a bunch of peripheral projects to this.
- Show replies
-
-
-
Money that would have been better spent on public transit and high speed rail.
-
Even if it's $100bn that only buys one California high speed rail. If that project ever even gets off the ground.
End of conversation
New conversation -
-
-
Hard to know what China is spending, Thiel is right AI is military tech, but wrong to think only that, huge upside for humanity if done right, big if
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
I believe it's much less than $100bn.
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
You cant have good object / scenario identification until you have good AGI. There is no shortcut. Only the data equivalent of a life of experience warns you that a cyclist might fall in your path because of a banana peel ahead on the road.
-
P(cyclist falls) almost exactly equals P(cyclist falls | banana peel in road that human driving car notices). Both humans and AI are best served by assigning a high probability of "will do random shit" to cyclists and leaving wide safety margins, regardless of the reasons.
End of conversation
New conversation -
-
-
Any new technology is nontransparent fractal. Finish the first visible sequence of tasks, uncover the next sequence. Would be great to know the way to assess fractal dimension of any new technology at the planning stage. Alas, its a gamble against rigged roulette
Thanks. 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.