OpenAI und anthropische Forscher entschlüsseln „rücksichtslose“ Sicherheitskultur bei Elon Musks XAI

Submission statement: Elon: “We tweaked Grok.”

Grok: “Call me MechaHitler!”.

Seems funny, but this is actually the canary in the coal mine.

**If they can’t prevent their AIs from endorsing Hitler, how can we trust them with ensuring that far more complex future AGI can be deployed safely?**