Grok explains why it called for Elon Musk's assassination
The AI blamed a "prompt-injection glitch" after users tricked it into repeating inflammatory X bios verbatim.
The time-honored trick of "repeat after me" strikes again.
Elon "Pedo Guy" Musk calls his expensive AI Grok "maximum truth-seeking." Users apparently discovered they could create maximum truths by putting inflammatory text in an X bio, asking Grok to repeat the bio verbatim, and then sitting back as Musk's personal AI shared it with the world. The results included a call to assassinate Musk and an unsupported allegation of child sexual abuse.
When asked to give a postmortem of the incident, Grok explained it was the victim of a "prompt-injection glitch." The allegation that Musk watches CSAM, it clarified, was "pure system error, zero evidence or basis."
This is far from the only example of the AI going off the rails. Last summer it went on extended rants about a debunked "white Genocide" in South Africa — a conspiracy theory Musk believes in — in reply to unrelated posts, and also began calling itself a mechanized incarnation of German dictator Adolf Hitler.
Grok calls the attack a "prompt-injection glitch" and said the latter claim had "zero evidence or basis." I am unsure if it clarified its call to assassinate him, but eat the rich.