Obviously, even the “Mecha Hitler” meltdown was not enough to scare the Pentagon.
A few days after Xai’s Grok Chatbot produced a stream of anti-Semitistic and homophobic reactions, Elon Musk’s company announced it had secured a $200 million defense contract to develop and supply AI tools to federal workers.
The Department of Defense Chief Digital and Artificial Intelligence Office announced the contract on Monday, saying it was designed to enable the government to take advantage of all the latest AI tools, from battlefield operations to business needs.
“The adoption of AI is changing the department’s ability to support our fighters and maintain a strategic advantage over our enemies,” Dr. Doug Matty, Chief Digital and AI executive at CDAO, said in a statement. “Utilizing off-the-shelf solutions in an integrated functional approach will accelerate the use of advanced AI as part of intelligence, business and enterprise information systems as part of the essential tasks of joint missions in the Warfighting domain.”
According to Lucas Hansen, co-founder of AI Security and Education’s nonprofit Civai, the Xai deal is about immediate use and maintaining pace with its global rivals.
“The government wants to support the development of frontier racebo and AI so that the US does not fall into China,” Hansen said. Decryption. “Then there’s the question of how those funds are distributed and how that support is provided. This may be the specific way you choose to support the company rather than expecting an internal deployment soon.”
However, Hansen warned that Grok’s underlying actions raises deeper concerns about how it will act in a high-stakes government setting.
“Even if Grok was deployed by the government using the same version that previously demonstrated his willingness to generate harmful responses, it is likely that he would behave radically unethical and autonomously, even if those use cases were not explicitly anti-Semitism or hatred,” he said.
Grok has been at the heart of several public meltdowns. In May, reports surfaced that Grok had inserted his “white genocide” claims at an unrelated prompt. Xai has condemned the problem of fraudulent employees. Last week, just before the release of Grok 4, the chatbot again sparked criticism of a wave of racist and homophobic output, leading to X CEO Linda Yaccarino’s resignation.
The concern remains about how Grok behaves in sensitive environments, but Hansen said the version the Pentagon uses is not the same as the version X users have accessed since November 2023.
The $200 million award is the latest deal between the US military and top AI developers. Other AI developers receiving the $200 million deal include Google, Anthropic, and Openai. In November, Meta said the open source llama AI model would be available to US defense agencies and contractors. In June, ChatGpt developer Openai was awarded a unique $200 million contract with the Department of Defense to launch the AI suite, Openai for Government.
If the Pentagon deployed Grok to its rival, Hansen emphasized that Grok is “openweight”, meaning that it can run on private hardware and does not require an API connection back to Xai.
Grok’s Announcement for the Government – A set of products that will enable US government customers to use frontier models
We are particularly excited about two new partnerships for US government partners
1) New contract from the US Department of Defense
2) We…– Xai (@xai) July 14, 2025
Hansen said despite Grok’s recent controversy, despite being at the heart of the trade, they will remain close to the key players shaping the future of AI, Hansen said.
“Most of this is a government that wants to have connections with all frontier rabos, and I think it really hit four important labs,” he said.
In another statement announcing its contract with the Pentagon, Xai said, “It’s about creating AI tools to support humanity in search of understanding and knowledge. To do this, Xai announced the launch of Grok for Government, a suite of AI tools, including the recently launched Grok 4 and Deep Search.
“America is a global leader in AI, and this is just a little less because of the tradition of innovation and a strong investment in engineering and science,” writes Xai. “We look forward to being able to contribute to the country that has made Xai unique.”
Xai did not respond to requests for comment Decryption.
