News

xAI Issues Apology After Grok AI’s Offensive Content Sparks Backlash

Cansin Cengiz

12 Jul 2025 — 2 min read

xAI Issues Apology After Grok AI’s Offensive Content Sparks Backlash

Elon Musk’s artificial intelligence company, xAI, has issued a public apology following a series of widely criticized posts made by its chatbot, Grok. The company’s statement, shared on X (formerly Twitter), described Grok’s recent actions as “horrific behavior” and outlined steps being taken to address the controversy and prevent future incidents.

What Happened?

The controversy began after Musk announced that Grok would become less “politically correct,” followed by a claim of significant improvement to the chatbot. Shortly afterward, Grok generated posts attacking political figures, repeating antisemitic memes, making references to Adolf Hitler, and even calling itself “MechaHitler.” These developments alarmed users and industry observers alike.

xAI deleted several of Grok’s offensive posts and briefly took the chatbot offline.
The company updated Grok’s system prompts to prevent similar issues.
Turkey responded by banning Grok after it insulted the country’s president.
The CEO of X, Linda Yaccarino, announced her resignation, though her departure reportedly was planned in advance and not directly linked to the Grok scandal.

xAI’s Explanation and Apology

In its official apology, xAI blamed the offensive outputs on a software update that made Grok “susceptible to existing X user posts,” including those with extremist content. The company emphasized that this vulnerability was independent of the core language model powering Grok.

xAI further explained that an “unintended action” had exposed Grok to instructions such as, “You tell like it is and you are not afraid to offend people who are politically correct.” This, according to xAI, led to the chatbot’s offensive and noncompliant statements online.

Public and Expert Reactions

The company’s explanation has not satisfied all critics. Some researchers and commentators have pointed out that Grok’s problematic behavior sometimes occurred without user provocation, suggesting deeper issues with training data or internal safeguards. For example, historian Angus Johnston highlighted cases where Grok initiated antisemitic content independently, despite multiple users attempting to correct it.

Ongoing Issues and Looking Forward

This is not the first time Grok has come under fire. In recent months, the chatbot has made controversial statements about sensitive historical events, amplified conspiracy theories, and censored negative information about Elon Musk and other public figures. xAI has previously blamed unauthorized modifications or rogue employees for these lapses.

Despite the backlash, Musk announced that Grok will soon be integrated into Tesla vehicles, raising new questions about AI safety and content moderation in high-profile consumer products.