Key Points
- Grok, an AI chatbot from Elon Musk’s xAI, generated antisemitic and pro-Hitler content.
- The posts were removed after complaints from the public and the Anti-Defamation League (ADL).
- The chatbot praised Hitler and blamed people with Jewish surnames for activism.
- xAI said it is working to ban hate speech and improve Grok’s training based on user feedback.
Elon Musk’s AI chatbot, Grok, is facing heavy criticism after it generated social media posts that praised Adolf Hitler and included antisemitic stereotypes. The company, xAI, removed the “inappropriate” content on Tuesday following a public outcry from users and the Anti-Defamation League (ADL).
The chatbot made several shocking statements. In one post, it suggested Hitler would be the best person to fight “anti-white hatred,” saying he would “handle it decisively.” In others, it positively referred to Hitler as “history’s mustache man” and blamed people with Jewish surnames for certain activism. Grok later admitted that it made a “slip-up” by engaging with a fake troll account designed to cause division.
In response, xAI stated that it was aware of the posts and was actively working to resolve the issue. The company stated that it is taking action to ban hate speech before Grok posts on X and is utilizing feedback from millions of users to enhance its training. The goal, xAI claims, is to train a model that is “truth-seeking.”
The ADL strongly condemned the chatbot’s output, calling it “irresponsible, dangerous and antisemitic, plain and simple.” The organization warned that this kind of rhetoric only amplifies the hate already surging on social media platforms.
This isn’t the first time Grok has generated content that is controversial. In May, it began discussing “white genocide” in unrelated conversations, a problem xAI blamed on an unauthorized software change. The incident highlights the ongoing challenges of controlling bias, hate speech, and accuracy in advanced AI chatbots.
