Will the U.S. military have a “spicy mode?” That’s a question that may arise as the U.S. armed forces adopt Elon Musk’s Grok as its artificial intelligence of choice. The decision comes as Grok AI becomes the target of a worldwide backlash against the proliferation of nonconsensual, sexualized images of women and children produced on the platform. The scale of the nudification problem appears to be large. Irish authorities, for example, said they are investigating 200 instances of child sex-abuse images generated by xAI chatbot Grok.
Grok already has been blocked in several countries and is facing increasing scrutiny from regulatory bodies around the globe. After initially dismissing concerns as an attempt at censorship, Elon Musk now says Grok will no longer be able to edit photos of real people to show them in revealing clothing in jurisdictions where it is illegal.
“We now geoblock the ability of all users to generate images of real people in bikinis, underwear and similar attire via the Grok account and in Grok in X in those jurisdictions where it is illegal,” said X in a statement. X added that only paid users on its social media platform will be able to edit images using Grok. Musk says this policy conforms to a “de facto standard in America” consistent with what can be seen in R-rated films, a notion that would make Grok the first AI with an “R” rating. Grok previously has ventured into uncharted hallucinatory meltdown territory, most memorably in 2025 with random anti-Semitic posts often characterized by violent graphics.
Critics quickly noted that virtual private networks (VPN) offer an easy work-around to Musk’s new guidelines. There are also questions about how Grok AI would distinguish between a real person and a fake one as well as what actions will be taken against rule breakers. Musk’s new directive also comes too late for the thousands of women and children who have already been victimized and whose images remain online.
The controversy surrounding Grok has not stopped the U.S. Department of Defense (aka War Department) from embracing it. DoD Secretary Pete Hegseth says Grok will be integrated into the U.S. military as part of an AI acceleration strategy, one that is notable for its absence of ethical considerations in favor of “hard-nosed realism.”
“Diversity, equity and inclusion and social ideology have no place in the DoW so we must not employ AI models which incorporate ideological “tuning” that interferes with their ability to provide truthful responses to user prompts,” according to documents posted on the department’s website. In other words, the Pentagon’s AI “will not be woke,” said Hegseth during an introductory speech at Musk’s SpaceX facility in Texas. Critics say Grok is an ideological match to the Trump administration, mirroring its darkest impulses and inclination to violate international law. “Meaningful human control” is to be determined by the War Department.
How well Grok performs and behaves is now under the purview of AWS veteran Cameron Stanley who assumes the mantle of the new position of chief digital and artificial intelligence officer. Hegseth also announced a “barrier removal SWAT team” at the Pentagon targeting “naysayers” who might impede AI development by voicing concerns over issues like data sharing.
The Pentagon’s AI acceleration strategy encompasses a number of initiatives:
— Swamp Forge is a competitive mechanism to discover ways of fighting with and against AI-enabled capabilities;
–Agent Network means to unleash AI agent development for battle management and decision making;
–Ender’s Foundry is to foster AI-enabled simulation capabilities (a name seemingly derived from the futuristic Ender’s Game military science fiction novels by Orson Scott Card).
Other initiatives include enhanced intelligence gathering and scenario planning, accelerated weapons development, AI agent development, open architecture use, and widespread access to Grok and Google’s Gemini AI. Separately, the U.S. military is expanding an AI initiative for mental health called BRAVE (Behavioral Health Resources and Virtual Experience). Given this close partnership between the Pentagon and Musk, it’s likely future AI developments will include a military role for Optimus, Musk’s humanoid robot.
But the real emphasis is on speed. “Speed defines victory in the AI era and the War Department will match the velocity of America’s AI industry,” says Emil Michael, undersecretary of war for research and engineering. “We’re pulling in the best talent, the most cutting-edge technology and embedding the top frontier AI models into the workforce—all at a rapid wartime pace.”
Just how fast the Pentagon turns AI into a technological door kicker may depend on hallucination rates. Musk points to a December, 2025 study by casino games aggregator Relum which identified Grok as having the lowest hallucination rate of 8% among 10 AIs tested and well ahead of ChatGPT at 35%. The challenge is that in military applications, an AI hallucination may result in human casualties and bring a new AI element to the term “collateral damage.”
