The Trump administration has asked the General Services Administration (GSA) to add xAI’s Grok chatbot to a list of approved vendors as soon as possible, WIRED magazine first reported after obtaining email correspondence regarding the issue on August 29.
“Team: Grok/xAI needs to go back on the schedule ASAP per the WH,” the email, from the commissioner of the Federal Acquisition Service, Josh Gruenbaum, allegedly urged.
Per WIRED, the request was made after a previous planned partnership with xAI disintegrated when Grok began spouting antisemitic rhetoric in July. For several days thereafter, Grok, which is integrated into X/Twitter, began responding to user prompts with antisemitic remarks, occasionally repeating Nazi propaganda and referring to itself as “MechaHitler.”
In June, xAI and the GSA met to discuss how the government could use Grok, but plans to this end seemed to fade after the July incident. Yet, according to reports, Grok 3 and Grok 4 both currently appear on the GSA Global Supply catalog, where numerous agencies can purchase items.
American media noted that this will allow federal workers to use Grok for tasks such as policy drafting, which could raise issues given the AI service’s erratic history.
Usage of Grok to spread malicious links
Then there is the issue of cybersecurity. On Thursday, cyber expert Nati Tal warned that cybercriminals, aided by Grok, were using a technique called “Grokking” to bypass X’s malvertising protections and spread malicious links.
“A malicious link that X explicitly prohibits in ads (and should have been blocked entirely) suddenly appears in a post by the system-trusted Grok account, sitting under a viral promoted thread and spreading straight into millions of feeds and search results,” Tal wrote on X.
The Jerusalem Post has reached out to the GSA for comment.