Democrats on the House Oversight Committee sent a letter to the General Services Administration on Friday demanding more information about how the agency is using Grok, the artificial intelligence chatbot built by Elon Musk’s xAI.
The correspondence comes after FedScoop reporting earlier this month revealed that government coders at GSA seemed to be looking at integrating Grok into their artificial intelligence work. Other sources told FedScoop that Grok had recently been approved for integration as an option into the GSAi app, a platform the agency has built to help federal workers access various generative AI models.
Four days after the publication of FedScoop’s story, xAI officially announced a “Grok for Government” service and confirmed that the company had been working to make its product available through GSA. As a result, Grok said ”every federal government department, agency, or office” could now access the company’s tools. xAI also announced a $200 million Defense Department contract.
The federal government’s interest in using Grok — which recently espoused antisemitic and pro-Hitler content — has received pushback from Democrats. A group of Jewish Democrats recently wrote to Defense Secretary Pete Hegseth about their concerns with the tool. Democrats in the House AI Caucus have also raised issues with the use of Grok, as has Senate Minority Leader Chuck Schumer, who condemned the Pentagon contract on the chamber’s floor.
But the latest letter, obtained by FedScoop, demands more information on the GSA’s work with Grok. The letter was addressed to Stephen Ehikian, the deputy GSA administrator who led the agency on an acting basis until earlier this week, and signed by Reps. Robert Garcia of California and Stephen Lynch of Massachusetts, the current and former ranking members of the committee, respectively.
“Alarmingly, it appears that federal agencies may be permitted to adopt Grok without following legally mandated procedures that protect privacy and cybersecurity,” the letter states. “The federal government should not be fast-tracking the adoption of technology with a demonstrably inaccurate and offensive bias that stands to financially benefit Mr. Musk, and doing so without the proper cybersecurity and privacy controls.”
The lawmakers flagged concerns about both the Cybersecurity Maturity Model Certification and FedRAMP, a program meant to help coordinate security reviews for cloud systems across federal agencies. Lawmakers also expressed interest in the public GitHub page, and the fact that it was taken down after FedScoop’s reporting. It’s not currently clear how GSA was approaching FedRAMP, but software or AI tools don’t inherently need to go through the FedRAMP process, as FedScoop has previously reported.
The lawmakers are asking for a series of records, including communications between xAI and GSA, communications between GSA and the Defense Department, and a full commit history of the GitHub repository that was taken down. The letter also requests a complete list of people based at GSA “involved in evaluating xAI models for use by the federal government and those individuals’ financial disclosures.”
This story was updated to reflect that the letter seeks communications between DOD and GSA.
Written by Rebecca Heilweil
Rebecca Heilweil is an investigative reporter for FedScoop. She writes about the intersection of government, tech policy, and emerging technologies.
Previously she was a reporter at Vox’s tech site, Recode. She’s also written for Slate, Wired, the Wall Street Journal, and other publications.
You can reach her at rebecca.heilweil@fedscoop.com. Message her if you’d like to chat on Signal.