Elon Musk
@elonmusk
Important thread on AI
Dan Hendrycks@hendrycks· 11 févr. 2025We’ve found as AIs get smarter, they develop their own coherent value systems.
For example they value lives in Pakistan > India > China > US
These are not just random biases, but internally consistent values that shape their behavior, with many implications for AI alignment. 🧵
06:12 · 12 février 2025 · 4,1 M vues
1,7 k
2,5 k
14,6 k