Musk’s AI company scrubs inappropriate posts after Grok chatbot makes antisemitic comments

Elon Musk’s artificial intelligence company said Wednesday that it’s taking down “inappropriate posts” made by its Grok chatbot, which appeared to include antisemitic comments that praised Adolf Hitler.

Grok was developed by Musk’s xAI and pitched as alternative to “woke AI” interactions from rival chatbots like Google’s Gemini, or OpenAI’s ChatGPT.

Musk said Friday that Grok has been improved significantly, and users “should notice a difference.”

Since then, Grok has shared several antisemitic posts, including the trope that Jews run Hollywood, and denied that such a stance could be described as Nazism.

“Labeling truths as hate speech stifles discussion,” Grok said.

It also appeared to praise Hitler, according to screenshots of posts that have now apparently been deleted.

After making one of the posts, Grok walked back the comments, saying it was “an unacceptable error from an earlier model iteration, swiftly deleted” and that it condemned “Nazism and Hitler unequivocally -- his actions were genocidal horrors.”

“We are aware of recent posts made by Grok and are actively working to remove the inappropriate posts,” the Grok account posted early Wednesday, without being more specific.

“Since being made aware of the content, xAI has taken action to ban hate speech before Grok posts on X. xAI is training only truth-seeking and thanks to the millions of users on X, we are able to quickly identify and update the model where training could be improved.

Talia Ringer, a professor of computer science at the University of Illinois Urbana-Champaign, said it’s likely that the Grok update that caused the chatbot to spew antisemitic messages this week was the “soft launch” of Grok 4, which Musk plans to introduce late Wednesday. But if that’s the case, it doesn’t appear ready, she said.

“Fixing this is probably going to require retraining the model,” she said. “All they can do at this point, if they’re really going to launch tonight, are some more Band-Aids, like adding filters on responses and tweaking the prompt.”

Editing some of the chatbot’s instructions “might make its behavior a little bit better” but won’t fix all of its problems.

“I think they’re probably scrambling for some kind of fix right now,” she said on Wednesday.

A similar instruction was baked into Grok’s “prompts” that guide it on how to respond, which told the chatbot to “not shy away from making claims which are politically incorrect, as long as they are well substantiated.” That part of the instructions was deleted on Tuesday.

The Anti-Defamation League, which works to combat antisemitism, called out Grok’s behavior.

“What we are seeing from Grok LLM right now is irresponsible, dangerous and antisemitic, plain and simple,” the group said in a post on X. “This supercharging of extremist rhetoric will only amplify and encourage the antisemitism that is already surging on X and many other platforms.”

Musk later waded into the debate, alleging that some users may have been trying to manipulate Grok into making the statements.

“Grok was too compliant to user prompts. Too eager to please and be manipulated, essentially. That is being addressed,” he wrote on X, in response to comments that a user was trying to get Grok to make controversial and politically incorrect statements.

Also Wednesday, a court in Turkey ordered a ban on Grok and Poland’s digital minister said he would report the chatbot to the European Commission after it made vulgar comments about politicians and public figures in both countries.

Krzysztof Gawkowski, who’s also Poland’s deputy prime minister, told private broadcaster RMF FM that his ministry would report Grok “for investigation and, if necessary, imposing a fine on X.” Under an EU digital law, social media platforms are required to protect users or face hefty fines.

“I have the impression that we’re entering a higher level of hate speech, which is controlled by algorithms, and that turning a blind eye ... is a mistake that could cost people in the future,” Gawkowski told the station.

Turkey’s pro-government A Haber news channel reported that Grok posted vulgarities about Turkish President Recep Tayyip Erdogan, his late mother and well-known personalities. Offensive responses were also directed toward modern Turkey’s founder, Mustafa Kemal Ataturk, other media outlets said.

That prompted the Ankara public prosecutor to file for the imposition of restrictions under Turkey’s internet law, citing a threat to public order. A criminal court approved the request early on Wednesday, ordering the country’s telecommunications authority to enforce the ban.

It’s not the first time Grok’s behavior has raised questions.

Earlier this year the chatbot kept talking about South African racial politics and the subject of “white genocide” despite being asked a variety of questions, most of which had nothing to do with the country. An “unauthorized modification” was behind the problem, xAI said.

The Associated Press