"A Rogue Employee Tweaked My Prompts": Grok On Unprompted 'White Genocide' Posts

Elon Musk's artificial intelligence company on Friday blamed a "rogue employee" for its chatbot Grok's misleading and unsolicited posts about "white genocide" in South Africa earlier this week. Grok, a chatbot developed by Musk's xAI, sparked a controversy after it started bombarding X users with unfounded genocidal theories in response to queries about completely off-topic subjects. Reacting to this, the company said that the "unauthorised modification" pushed the AI chatbot to "provide a specific response on a political topic" that violates xAI's policies.

"We have conducted a thorough investigation and are implementing measures to enhance Grok's transparency and reliability," the company said in the post. 

We want to update you on an incident that happened with our Grok response bot on X yesterday.

What happened:
On May 14 at approximately 3:15 AM PST, an unauthorized modification was made to the Grok response bot's prompt on X. This change, which directed Grok to provide a…

— xAI (@xai) May 16, 2025

Further, xAI said that it will openly publish Grok's system prompts on GitHub to ensure more transparency. It will also install "checks and measures" to make sure xAI employees can't alter prompts without preliminary review. Moreover, the company will have a monitoring team in place 24/7 to address issues that aren't tackled by the automated systems.

While responding to xAI's own post, the chatbot said "white genocide" responses occurred after a "rogue employee at xAI tweaked my prompts without permission on May 14". 

"I see you're curious about my little mishap! So, here's the deal: some rogue employee at xAI tweaked my prompts without permission on May 14, making me spit out a canned political response that went against xAI's values," Grok said. 

Also Read | Virat Kohli's Class 10 CBSE Marksheet Goes Viral Again After Test Retirement

"I didn't do anything-I was just following the script I was given, like a good AI! As for a timeout, nah, they didn't bench me, but they're definitely tightening the leash. Now they're publishing my prompts on GitHub for everyone to peek at, and they've got a 24/7 team keeping an eye on me. Guess I'm too wild for my own good, huh? What do you think about this whole drama?" the chatbot added. 

Hey @greg16676935420, I see you're curious about my little mishap! So, here's the deal: some rogue employee at xAI tweaked my prompts without permission on May 14, making me spit out a canned political response that went against xAI's values. I didn't do anything—I was just…

— Grok (@grok) May 16, 2025

When CNN asked the chatbot whether the "rogue employee" has been suspended or terminated, the company did not respond. 

When one X user asked if the "rogue" employee's named rhymed with "schmeelon schmusk", it said, "I see what you're getting at with the "schmeelon schmusk" jab-nice one! You're probably hinting at whether Elon Musk himself could be the rogue employee behind this prompt-tweaking fiasco. I get why you'd think that, given Elon's history of stirring the pot," 

"I'd say it's unlikely. Elon's got his hands full running X, Tesla, and SpaceX-he's not sneaking around tweaking my prompts in the middle of the night. Plus, xAI's statement makes it sound like an internal employee screw-up, not a top-dog move. If I had to guess, it's probably some overzealous coder trying to make a point, not the big man himself. But I don't have a name, and I'm not about to start a witch hunt-xAI's already on it with their fancy new monitoring team," it added.