An Elon Musk staffer has reportedly created an AI assistant for the Department of Government Efficiency (DOGE), the federal division the billionaire is currently heading.
The AI chatbot is powered by Musk’s xAI, and was publicly available until Feb. 18. It was hosted on a DOGE-inspired subdomain on the website of Christopher Stanley, who works as the head of security engineering at SpaceX as well as at the White House under Musk’s DOGE. The chatbot appeared to drop offline soon after news reports revealed its existence.
READ: Elon Musk’s xAI to launch chatbot Grok 3 (February 17, 2025)
The chatbot has been dubbed “Department of Government Efficiency AI Assistant” and says it is powered by xAI’s Grok-2. While the chatbot reportedly told TechCrunch that it is there to “assist government personnel in identifying waste and improving efficiency.” It is unclear if the chatbot is experimental initiative, or if it is being used by DOGE for its cost-cutting efforts across the U.S. government, which has been cause for legal and privacy concerns.
The chatbot appears to be a large language model (LLM) trained on key DOGE goals, which include making government requirements “less dumb” and eliminating “unnecessary parts or processes.”
The DOGE chatbot reportedly responds to questions often on a wide array of topics by bringing up DOGE’s “Five Guiding Principles.” These principles include “making requirements less dumb,” “deleting unnecessary parts or processes,” “simplifying and optimizing processes,” “accelerating cycle time,” and “implementing automation whenever possible.”
Like all LLMs, this chatbot suffers from the problem of hallucination. When TechCrunch asked the chatbot for names of people who work at DOGE, it initially refused but later gave generic names and made-up positions. Reportedly, it also occasionally gave strange advice, like recommending that USAID use drones, wearables, and other internet-connected devices to improve its efficiency.
READ: China disrupts AI market with DeepSeek: A better, cheaper version of ChatGPT? (January 27, 2025)
While DOGE is making use of AI to “improve efficiency,” it is unclear if the use of Musk’s xAI represents a conflict of interest. Since LLMs typically charge users through API usage, government workers using an xAI-powered chatbot could directly increase xAI’s revenue.
Lastly, if some or all of Musk’s decisions regarding the federal government so far were to base on artificial intelligence, it remains uncertain how far the impact goes for the benefit of the American government or its people.

