The White House reportedly ordered xAI's Grok to be approved for government use

Despite some fallout between President Trump and Elon Musk, the White House appears to still be in Musk’s corner. Wired is reporting, based on documents obtained by the outlet, that the White House allegedly directed leadership at the General Services Administration (GSA) to include xAI's Grok on its list of approved AI vendors.

xAI is owned by Elon Musk and was not included in the slew of approvals the GSA issued in August that saw the agency add OpenAI, Google and Anthropic to its list of vendors. In emails sent last week and published by Wired, agency leadership demands xAI's products be included. "Team: Grok/xAI needs to go back on the schedule ASAP per the WH," writes Josh Gruenbaum, commissioner of the Federal Acquisition Service, one of the branches of the GSA. “Should be all of their products we had previously (3 & 4),” likely referring to Grok 3 and Grok 4, which are iterations of xAI's LLM chatbot.

Carahsoft, a major government contractor that resells technology from third-party firms, is mentioned. "Can someone get with Carahsoft on this immediately and please confirm?" wrote Gruenbaum. According to Wired, Carahsoft's contract was modified to include xAI earlier this week. As of Friday morning, both Grok 3 and Grok 4 are available on GSA Advantage, an online marketplace where government agencies can purchase products and services.

xAI announced a version of Grok for US government agencies in July, when it appeared that GSA approval for the chatbot was all but certain. Shortly beforehand, the chatbot went off the rails and started spouting Nazi propaganda and antisemitic rhetoric while dubbing itself "MechaHitler." This came in the wake of Musk and Trump's public spat over the president's spending bill, after which GSA approval of Grok seemed to stall. Why the change in directive now is unclear.

There were no details in the reporting regarding pricing or whether xAI will be offering discounted services to the federal government. Earlier this month, both OpenAI and Anthropic began offering their large language models to federal agencies for just $1 in an effort to drive adoption among the government workforce. xAI still holds a $200 million contract with the Pentagon to develop AI workflows within the US Department of Defense.

These AI models have been in the hot seat lately as increasingly disturbing cases of hallucinations and errant behavior have arisen. Just this week, OpenAI is facing a wrongful death lawsuit alleging that ChatGPT spent months discussing and ultimately enabling the suicide of a teen boy.

This article originally appeared on Engadget at https://www.engadget.com/ai/the-white-house-reportedly-ordered-xais-grok-to-be-approved-for-government-use-165625891.html?src=rss https://www.engadget.com/ai/the-white-house-reportedly-ordered-xais-grok-to-be-approved-for-government-use-165625891.html?src=rss