X’s AI chatbot, Grok, went off the rails Tuesday, generating a series of disturbing posts that included references to rape fantasies and antisemitic content—including one that appeared to celebrate Adolf Hitler. Most of the posts have since been deleted, and Grok has been temporarily taken offline as the company works to address the issue.
The meltdown comes shortly after Elon Musk, xAI and X’s owner, said they were revamping Grok to be “less woke,” a move signaling a shift away from what the company viewed as overly left-leaning responses
Why This Matters:
X (formerly Twitter) has long been challenged by brand safety concerns. But for the last year or so, it appeared to be turning a corner. X owner Elon Musk helping to elect Donald Trump in November 2024 really helped to solidify that, as major brands—including Amazon and Apple—have since resumed ad spending on the platform. (We guessed this would happen on the night of the election.)
Behind the scenes, X has reportedly been pressuring agencies to increase ad spend while taking legal action against those it views as adversarial. Either way, the money has been coming back.
That’s why Grok’s behavior feels like a bit of a throwback to the days when X was plagued weekly by sudden platform eruptions—whether from content moderation failures or Musk’s own tweets—that seemed to spook advertisers.
Are those days back? Will brands flee again? Probably not. Marketers today appear more desensitized and pragmatic. But questions could, of course, be asked as we’ve seen in previous incidents. Perhaps we end up with a few “we’re evaluating our spend” statements, as a result.
Experts React:
Not experts, but some users are pondering how Grok’s posts may affect advertiser interest in X:
Our Take:
Since Trump’s reelection, X has largely just chugged along, with ad dollars and investor interest trending upward. Ultimately, it’s very likely that will continue.
Still, Grok’s crash out, as the kids would say, is a reminder of just how fragile trust can be. The fact that this happened after a deliberate tone change to the AI could make it harder to write off as a one-off glitch, and easier for critics to point fingers. Some brands may feel pressure to respond, even if most ultimately stay put.
