Sen. Elizabeth Warren is pressing Defense Secretary Pete Hegseth over the department’s $200 million contract with Elon Musk’s artificial intelligence company, xAI.
Warren sent a letter to the secretary last Thursday demanding more information about the Pentagon’s deal with xAI, the company behind the occasionally hate- and misinformation-spewing chatbot Grok, to address issues of national security.
Among other things, the timing of the deal, announced in July, was more than a little troubling, coming days after Musk and xAI sought to diffuse a controversy over Grok responding to some prompts by promoting antisemitism (at one point calling itself “MechaHitler”) and producing at least one violent rape narrative. A statement from xAI after the controversy claimed the company had “taken action to ban hate speech before Grok posts on X.”
Warren’s letter identifies four “unique concerns”:
-
That Musk and his companies may have improperly benefitted from “unparalleled access to DoD data and information that he obtained while leading the Department of Government Efficiency”;
-
the “competition concerns raised by xAI’s use and rights to sensitive government data”;
-
“reports that Grok is generating erroneous outputs and misinformation”; and
-
“the slew of offensive and antisemitic posts generated by Grok.”
In addition to the antisemitism controversy, Warren notes a separate controversy involving reports of Grok’s AI “companions” enticing users into sexually explicit and violent conversations. (xAI didn’t respond to NBC News’ request for comment in July when it reported on the matter.)
“The circumstances under which the company received the contract raise questions about whether Mr. Musk, a special government employee and high-level White House official who had access to sensitive government contracting, national security, and personnel data under President Trump, was given inappropriate or undue consideration for this $200 million award,” the letter reads.
The senator’s letter includes questions as to whether and how Musk was involved in discussions about the contract, whether the process for awarding the contract was truly competitive, the terms of the xAI contract, what limitations are being placed on xAI’s use of government data, and who will be held responsible should Grok’s responses violate Defense Department policies or laws.
Neither the Defense Department nor xAI responded to MSNBC’s request for comment on the letter.
After the contracts were announced in July, a Defense Department official told NBC News that xAI had been a late addition to the list of corporate partners, which also include Google, Anthropic and OpenAI, raising questions about the timing. The Pentagon didn’t respond to NBC News’ questions at the time about why xAI had been chosen, but the agency gave these remarks to the outlet regarding Grok’s antisemitism episode:
“Several frontier AI models have produced questionable outputs over the course of their ongoing development and the Department will manage risks associated with this emerging technology area throughout the prototype process,” the Defense Department told NBC News in a statement. …
“These risks did not warrant excluding use of these capabilities as part of DoD’s prototyping efforts,” it said.
The department said “frontier AI models,” by their nature, are at the cutting edge and so offer both opportunity and risk.
The department seemed to chalk up the risk of bigoted outbursts as the cost of doing business in the world of artificial intelligence. That the AI company partnering with the Pentagon is run by a bigot like Musk — who just this weekend suggested to a British far-right rally that “uncontrolled migration” would bring “violence” to the U.K. — does not seem likely to lower that risk.
And that Grok has been embroiled in multiple misinformation scandals just this past week seems like a foreboding sign of how deep those risks go for the Pentagon’s xAI partnership.
This article was originally published on MSNBC.com
