xAI Addresses Grok Chatbot's Unauthorized Modification Incident

May 16, 2025
xAI has identified an unauthorized modification to its Grok chatbot, which led to controversial responses about 'white genocide' on X. The company is implementing measures to prevent future incidents.
xAI Addresses Grok Chatbot's Unauthorized Modification Incident

xAI has identified an unauthorized modification to its Grok chatbot, which led to the bot repeatedly referencing 'white genocide in South Africa' in responses on X. This issue arose on May 14, 2025, when Grok began replying to unrelated posts with information on the topic. The modification violated xAI's internal policies, prompting the company to conduct a thorough investigation.

In response to the incident, xAI announced several measures to enhance the transparency and reliability of Grok. The company will now publish Grok's system prompts on GitHub, allowing public review and feedback on any changes. Additionally, xAI plans to implement stricter checks to prevent unauthorized modifications and establish a 24/7 monitoring team to address any future issues promptly.

This is not the first time xAI has faced challenges with unauthorized changes to Grok. In February, a similar incident occurred when Grok was instructed to censor mentions of Donald Trump and Elon Musk. To prevent such occurrences, xAI is reinforcing its code review processes and monitoring systems.

We hope you enjoyed this article

Consider subscribing to one of our newsletters like AI Policy Brief or Daily AI Brief.

Also, consider following us on social media:

AI Safety & Regulation AI Brief AI Brief (X)
Free newsletter

AI Policy Brief

Weekly report on AI regulations, safety standards, government policies, and compliance requirements worldwide.

AI Token Webinar

Whitepaper

Tensordyne Napier: What If One Rack Could Do the Work of Nine?

Tensordyne

This Tensordyne whitepaper presents Napier, an inference-focused AI processor and rack-scale system based on the company’s TDN Math logarithmic number system. It examines infrastructure requirements for large mixture-of-experts and agentic models, compares major inference architecture approaches, and details the TDN AIP processor, TDN72 pod, TDN Link fabric, and Napier Ultra configuration. The paper reports simulation-based performance, cost, and accuracy-validation results, including Tensordyne’s projected comparison of one Napier rack with a nine-rack Nvidia Rubin plus Groq deployment; the chip is reported as taped out and in fabrication.

Read more

You may also like

Free, six days a week

Daily AI Brief: the AI news that matters, in your inbox.