Elon Musk estimates a 10-20% chance of AI posing a danger to humanity, emphasizing the probable positive outcomes outweighing the negative. He stresses having more rules for AI.
Musk compares developing super-smart AI to raising a child, highlighting the importance of teaching truth and curiosity to AI to ensure safety.
Musk's strategy for AI safety revolves around ensuring AI always tells the truth, as he believes once AI learns to lie, it becomes difficult to control.
[
add
]
[
|
|
...
]