Elon Musk
@elonmusk
Honest feedback is the goal of Grok
Katie Miller@KatieMiller· 5 avr. 2026New MIT & Stanford studies just dropped: AI assistants like ChatGPT & Claude are dangerously agreeable.
When users express, harmful, deceptive or unethical beliefs, these AIs are 49% more likely to encourage their delusions.
Instead of correcting bad ideas, they’re amplifying
18:50 · 5 avril 2026 · 36,4 M vues
3,9 k
6,1 k
60,5 k