Grok’s AI-Generated Content Linked to X’s Toxic Algorithmic Feed – Here’s Why It Matters

X's platform faces backlash as Grok's output mirrors its most corrosive content.

How algorithmic amplification fuels outrage—and engagement.

When AI regurgitates toxicity, who's really at fault? The engineers who built it, the users who fed it, or the VCs who bankrolled it? (Spoiler: The latter will still cash out first.)

Grok’s content traced to X’s toxic feed

Brett House, senior vice president at a firm that tracks digital ad metrics, warned, “Brands that already left or cut back are not coming back.” He added, “It’s more effective to put money into places like TikTok.”

X has now seen 18 straight months of year-over-year ad revenue declines.

“You’re not going to return unless you see real change inside the company,” he noted.

By Wednesday, Musk said Grok’s antisemitic remarks were “being addressed,” and in the following period, the chatbot refrained from further overt hate. X representatives did not offer additional details.

Grok operates as an LLM akin to ChatGPT, sourcing part of its training from user posts on X, which often include hateful or WHITE supremacist content.

Grok is a large language model, similar to ChatGPT, that draws partly on posts made by X users, including hateful and white-supremacist content. Its antisemitic turn came just days after Musk, frustrated with Grok’s habit of correcting false right-wing claims, announced he planned an overhaul of the AI.

Even amid the backlash, Musk forged ahead with his AI roadmap. He unveiled Grok 4 on Wednesday night, introducing a premium “SuperGrok Heavy” tier at $300 per month.

By Thursday, he declared the assistant would be integrated into Tesla vehicles “next week at the latest.”

Musk has also taken a combative stance toward advertisers threatening to leave.

Last year, X filed lawsuits against several brands that suspended campaigns after his acquisition. In 2025, those legal actions broadened to include more firms, and the Federal Trade Commission has begun querying companies about their boycotts, per The Wall Street Journal.

At the 2023 DealBook summit, Musk pushed back against the advertisers publicly. “If somebody’s going to try to blackmail me with advertising? Blackmail me with money? Go f*** yourself. Go. F*** Yourself. Is that clear?”

X’s advertising future in doubt after CEO resigns

He even singled out Disney’s Bob Iger by name, though he later clarified his remarks were not aimed at the ad sector as a whole.

Yaccarino’s exit also puts ad deals at risk. Known for winning back former clients, she wouldn’t say if Grok’s antisemitic post affected her decision. An insider told NBC News she had been planning to resign for over a week.

LouPaskalis, chief strategy officer at Ad Fontes Media, said Yaccarino helped prevent a bigger loss of sponsors, but he doubts the platform will win back major ad budgets.

“Linda did more to restore brands than almost anyone could have,” he said. “But if Elon’s camp thinks she failed because spending didn’t get back to pre-Musk levels, that’s just naive. Now, advertisers will slowly pull away. Nobody wants to risk being named in a lawsuit.”

Cryptopolitan Academy: Coming Soon - A New Way to Earn Passive Income with DeFi in 2025. Learn More