محاور الجلسة
أهداف النظر
1
If AI becomes exceptionally good at answering questions, who decides which questions deserve to be asked?
2
Can AI optimize institutional decisions when the objectives themselves involve values and trade-offs?
3
Is a human still genuinely in control if they cannot fully understand, challenge, or reconstruct the AI's reasoning?
Supporters