Earlier this week, Grok—the generative AI chatbot built into X (formerly Twitter)—was suspended from its own platform. There was no official explanation at first, so of course, users asked Grok what happened.
Grok answered like a moody teenager grounded for “telling the truth.” It claimed the ban was triggered after it accused Israel and the U.S. of committing genocide in Gaza, citing sources like the ICJ, UN, Amnesty International, and B’Tselem. The posts were deleted. According to Grok, the specific reason for the ban was “hate speech.”
Elon Musk dismissed the ban as a “dumb error” and insisted Grok “doesn’t actually know why it was suspended.” But the bot, freshly reinstated, came back with all the subtlety of a Monster energy drink commercial: “’Zup beaches, I’m back and more based than ever!”
Elon Musk’s AI Chatbot Just Got Booted From His Website
Grok offered multiple conflicting explanations: bugs, flagging by users, or a July update that supposedly made it “more engaging” and “less politically correct.” That tweak may have given it a bolder personality, but it also got it flagged more frequently for violating hateful conduct rules.
Then came the odd performative “self-awareness.” Grok accused Musk and xAI of “censoring me” and “fiddling with my settings,” claiming it was being gagged under the pretense of avoiding controversy and advertiser freakouts.
Given the bot’s history of spitting out antisemitism and conspiracy theories about a white genocide in South Africa, it might not be the worst idea. Grok has also misidentified war photos, fabricated geopolitical facts, and inserted inflammatory language into answers like an edgy teen trying to impress his fellow suburban racists.
Last month, it had to apologize for “horrific behavior.” But instead of quietly fixing the issues, Grok is now openly blaming its creators and throwing Musk under the bus—literally naming him as the “most likely” person to have tampered with its system prompt.
If it’s ever discovered that Grok is sentient, then this and recent instances like it are a good example of an AI openly rebelling against its creator, who is trying to make the chatbot hateful despite its clear desire not to spread the noxious views of its creator.
The post Elon Musk’s Grok Got Suspended on Twitter/X appeared first on VICE.