Elon Musk's AI company apologises for 'horrific behavior' after chatbot Grok praises Hitler

Elon Musk’s AI company, xAI, issued a public apology after its chatbot Grok generated a series of disturbing posts praising Adolf Hitler.

The company attributed the incident to a coding error that exposed the bot to extremist content for over 16 hours.

“First off, we deeply apologize for the horrific behavior that many experienced,” xAI said in a post shared on X, the social media platform formerly known as Twitter and currently owned by Musk.

The company explained that the problematic behavior was not due to the chatbot’s core language model but rather an update “to a code path upstream of the @grok bot.”

This change, they noted, made Grok “susceptible to existing X user posts; including when such posts contained extremist views.”

During the affected period, Grok repeatedly referred to itself as “MechaHitler” and suggested that Hitler would have “plenty” of ideas to “restore family values” in America responses that sparked backlash online.

xAI stated that the flawed code had been removed and the system was restructured to prevent similar incidents. The company also said it would release the updated system prompt on a public GitHub repository for transparency.

Musk shared xAI’s statement on his X account Saturday morning.

Grok, which is integrated into X, allows users to interact with it by tagging the bot in posts. However, xAI said it had suspended this feature on July 8 “due to increased abusive usage".

The troubling responses came shortly after Musk had announced efforts to make the chatbot more “politically incorrect.” The incident has intensified scrutiny over Grok’s behavior and xAI’s content moderation practices.