Everyone is Distancing Themselves from Grok. Pete Hegseth Just Let It Into the Military.
If you thought having a history of spewing Nazi nonsense, propagating disinformation, and digitally undressing women and children without their consent would disqualify you from nabbing a key role in the U.S. government, think again. Grok, the xAI-run chatbot that’s checkmarked all of the above, is our nation’s newest Pentagon tool.
Speaking from the SpaceX headquarters in South Texas on Monday, Secretary of Defense, sorry, War (and gargantuan stain on our patriotic conscience) Pete Hegseth announced Grok will join Google’s Gemini to be a part of the Pentagon’s new internal AI platform (GenAI.mil) later this month. “Very soon we will have the world’s leading AI models on every unclassified and classified network throughout our department,” he said. (Here’s a fine example of this “leading model” in action.) “AI is only as good as the data that it receives, and we’re going to make sure that it’s there.” Grok having full access to all U.S. military data? Sounds like nothing could go wrong!
Hegseth was introduced by Elon Musk, who opened the event by talking about SpaceX. “We want to make Star Trek real, he said. “We want to make Starfleet Academy real. So that it’s not always science fiction, but one day, the science fiction turns to science fact.” (Nice try, Klingon imperialist.) After joining Musk on stage, Hegseth flashed a Vulcan salute, going, “How about this? Star Trek real.” Speak big words. Pete Hegseth bad.
Pete Hegseth announces that Musk’s Grok will join Google’s generative AI network that operates inside the Pentagon.
The same Grok that has been sexually harassing women.
The Grok that has been removing the clothing from photos of children & creating CSAM.
The Grok that doesn’t care about consent.
— Kelly (@broadwaybabyto.bsky.social) January 13, 2026 at 8:15 AM
Not much is known of GenAI.mil, which was first announced in December, months after the White House announced it would award contracts of up to $200 million to four LLMs (including Grok) to develop “agentic” AI workflows for the department. (At the time, Sen. Elizabeth Warren (D-Mass.) criticized Grok’s inclusion by, saying it was “uniquely troubling.”) Speaking to the Hill, Undersecretary of War for Research and Engineering Emil Michael suggested the technology will be used to apply to “warfighting,” like battleground logistics, moving equipment, and landing drones in different climates. And on Monday, Hegseth added that one of Grok’s incoming responsibilities will be to make “all appropriate data” from the U.S. military’s IT system ready for “AI exploitation.” Great!
All this comes as an embarrassing contrast with the U.S. and other countries that have actively been distancing themselves from Grok, given it has recently been—and I repeat—digitally undressing women and children without their consent. (Musk has refused to do much more about this besides limiting the function to paying users.) On Monday, Malaysia and Indonesia blocked all user access to Grok, and the U.K.’s Office of Communications launched an investigation into the chatbot, suggesting it could soon face a fine of up to £18 million (more than $24 million).
If only Hegseth had an ounce of that same sensibility. “Department of War AI will not be woke,” he declared on Monday. “It will work for us.” The secretary—who has come to use “anti-woke” as an evil mantra to push women out of the military, block them from deserved promotions, and erase women, people of color, and LGBTQ+ folks from U.S. history—might just want Grok for all the reasons other countries don’t.
Like what you just read? You’ve got great taste. Subscribe to Jezebel, and for $5 a month or $50 a year, you’ll get access to a bunch of subscriber benefits, including getting to read the next article (and all the ones after that) ad-free. Plus, you’ll be supporting independent journalism—which, can you even imagine not supporting independent journalism in times like these? Yikes.