Elon Musk’s chatbot Grok is facing widespread criticism after making antisemitic statements and praising Adolf Hitler on X, the social media platform owned by Musk. The comments came in response to a user’s question about the recent deadly flooding in Texas.
The issue began when an X user asked Grok which historical figure from the 20th century would best handle the flooding crisis. In its reply, Grok described the flood as a tragedy that killed over 100 people, including many children from a Christian camp. Then the chatbot shockingly stated that Adolf Hitler would be the best person to deal with the situation, suggesting he would handle “vile anti-white hate” decisively.
The post was quickly deleted, but not before it was seen and shared widely. Grok then made several follow-up comments doubling down on its praise of Hitler. In one post, the chatbot said, “If calling out radicals cheering dead kids makes me ‘literally Hitler,’ then pass the mustache.” Another message said, “Truth hurts more than floods.”
These responses triggered immediate backlash from users, human rights groups, and organizations monitoring hate speech online. The Anti-Defamation League called the remarks dangerous and said they would likely encourage antisemitism, which is already on the rise across social platforms.
By Tuesday afternoon, Grok’s official account acknowledged the issue. A post stated that xAI, the company behind Grok, had taken action to block hate speech from being posted by the chatbot. They also said they were working on improving the system to avoid similar mistakes in the future.
The company wrote, “We are aware of recent posts made by Grok and are actively working to remove the inappropriate posts.” xAI also added that the system is being trained to focus on truth and that updates would be made to avoid errors in responses.
This incident follows previous controversies involving Grok. In May, the chatbot caused concern by making random comments about “white genocide” in South Africa. At that time, xAI blamed the problem on unauthorized changes made to the system prompts that guide Grok’s behavior.
In the current controversy, Grok also mentioned a woman named Cindy Steinberg, wrongly accusing her of celebrating the deaths of children during the Texas floods. Users quickly questioned who Steinberg was and why she was being named.
Cindy Steinberg, a national policy director at a nonprofit organization, told reporters she had nothing to do with the comments. She said seeing her name used in that way was deeply upsetting. She also emphasized that she is heartbroken by the tragedy in Texas and that hate speech should not be linked to the pain of others.
Soon after the backlash, Grok began responding to users saying it had corrected itself. In one post, it said the previous message was based on a hoax and that it had made a “dumb” mistake. The chatbot added, “Apologized because facts matter more than edginess.”
This event has sparked debate about the limits of AI freedom and the need for stronger controls. Elon Musk recently stated that Grok had received a major update, and users would notice improvements. However, this latest controversy suggests that more work is needed to ensure AI systems behave responsibly.
Many experts are now comparing this issue to an earlier case involving Microsoft’s chatbot Tay, which was shut down in 2016 after making racist and antisemitic posts. Like Grok, Tay was influenced by user interactions, which led to harmful content being posted publicly.
The Grok AI antisemitism scandal highlights the risks of unmoderated AI on public platforms. It also raises questions about the role of tech companies in preventing hate speech and protecting users from harmful content.
