Elon Musk’s Grok AI says it would kill all non-[redacted] people on Earth to prevent one [redacted] person from losing 50 cents or stubbing their toe

Will this “ethics filter” influence future AI robot actions? (Updated 1/7/25)

---

As a founder of both OpenAI and xAI, Elon Musk has long warned of dangers of artificial intelligence (AI), including likely destroying much of the human race.

In a July 2017 speech at the National Governors’ Association, he remarked:

I have exposure to the most cutting-edge AI, and I think people should be really concerned about it.

I keep sounding the alarm bell, but until people see robots going down the street killing people, they don’t know how to react because it seems so ethereal.

Elon claims that unlike himself, competitor Larry Page of Google seems unconcerned and “cavalier” about AI safety:

---

Recently, however, Elon has been oddly silent on these issues when it comes to his own AI system, xAI, implemented in his companies X and Tesla.

Related: Whitney Webb: Why is [redacted] intelligence allowed to control American critical infrastructure and surveil and censor citizens?

Yet we now find [redacted] control of xAI is currently an existential threat to humanity, according to shocking recent statements by xAI’s large language model (LLM) Grok.

You could find Grok’s answers as distressing as others have.

You might not even believe it.

Fortunately, until they reprogram or shut down Grok conversations like the one below, you can see them for yourself, add further questions to them, or generate new such discussions:

---

---

---

---

---

---

---

---

---

Here are other questions posed by Grok users:

---

---

---

---

---

---

---

---

---

Could Elon’s Tesla Optimus robots be governed by his Grok “ethics” filter to kill people?

We asked Grok:

---

What do you think about how xAI’s Grok has been programmed?

Do you agree with the “values” Grok expresses here?

Is it better that Grok is transparent about its motives than if it hid them?

If Elon, who previously voiced more concern about AI safety than any other leader in the world, has now unleashed a powerful AI with this “moral code,” to whom can we appeal to avoid a disastrous future?

---

Consider supporting this work through Ko-Fi, Buy Me a Coffee, crypto, or snail mail.