The United States' Defense Department has announced the award of a $200 million contract to xAI's Grok, the chatbot most recently in the news for declaring itself "MechaHitler" and going on bizarre rants about anti-White oppression. What could go wrong!
The news came alongside xAI's announcement of "Grok for Government," a toolset that lets agencies adapt the chatbot for specific purposes. xAI said its offerings are now "available to purchase via the General Services Administration (GSA) schedule" with "every federal government department, agency, or office" able to buy them.
Get ready for some high-level nonsense.
"The adoption of AI is transforming the Department’s ability to support our warfighters and maintain strategic advantage over our adversaries," said chief digital and AI officer Doug Matty. "Leveraging commercially available solutions into an integrated capabilities approach will accelerate the use of advanced AI as part of our joint mission-essential tasks in our warfighting domain as well as intelligence, business, and enterprise information systems.”
That phrase "in our warfighting domain" seems quite troubling, though what Grok is actually going to be used for remains intentionally vague. One would hope it's not used for anything that's, y'know, actually important or that lives depend upon: because like all LLMs Grok frequently gets things wrong and is an unthinking piece of software with no understanding of its actions or their consequences. I mean, you don't want this thing anywhere near decisions that actually matter.
Just last week Grok went on an antisemitic tear, and decided to tell people that Hitler had some good ideas, actually. Elon Musk claimed that the poor thing had been "manipulated" by nasty humans, an occupational hazard for any chatbot, and a day later xAI announced Grok 4, which it claims is "the most intelligent model in the world."
The deal comes after the Trump administration, of which Musk was a member until recently, passed an executive order in April promoting the adoption of AI. Trump also revoked a 2023 executive order issued by President Joe Biden intended to place various safeguards on the technology.
xAI joins OpenAI, Anthropic and Google in having a $200 million contract with the DoD, which is clearly keeping its options open. The agency says it will be partnering with the General Services Administration to make the tools available throughout the federal government.
"Under the umbrella of Grok For Government, we will be bringing all of our world-class AI tools to federal, local, state, and national security customers," said xAI in a statement. "These customers will be able to use the Grok family of products to accelerate America—from making everyday government services faster and more efficient to using AI to address unsolved problems in fundamental science and technology."
Oh yeah: Musk reckons that xAI is going to discover "new physics" and "new technologies" within the next year. Good luck with that.
The big question here is what Grok and the other LLMs are actually going to be used for. Grok itself tells me it'll be "analyzing data in real time to flag threats, such as detecting anomalies in network traffic for cybersecurity or identifying patterns in surveillance data."
That likely means processing satellite imagery and signals intelligence, though actual decision-making will remain in the hands of humans. DoD directive 3000.09 on "autonomy in weapon systems" emphasises that humans must be involved in any decision leading to potentially lethal actions. So relax: Grok can't nuke us. For now, anyway.
