The Pentagon announced Monday it is going to spend almost $1 billion on “agentic AI workflows” from four “frontier AI” companies, including Elon Musk’s xAI, whose flagship Grok appeared to still be declaring itself “MechaHitler” as late as Monday afternoon.
In a press release, the Defense Department’s Chief Digital and Artificial Intelligence Office — or CDAO — said it will cut checks of up to $200 million each to tech giants Anthropic, Google, OpenAI and Musk’s xAI to work on:
- “critical national security challenges;”
- “joint mission essential tasks in our warfighting domain;”
- “DoD use cases.”
The release did not expand on what any of that means or how AI might help. Task & Purpose reached out to the Pentagon for details on what these AI agents may soon be doing and asked specifically if the contracts would include control of live weapons systems or classified information.
A Defense official responded, saying that “Awards to Anthropic, Google, OpenAI, and xAI will enable the Department to leverage the technology and talent of U.S. frontier AI companies to develop agentic AI workflows across a variety of mission areas, including warfighting, intelligence, business, and enterprise information systems.”
That term, “Frontier AI,” the official said, referred to “companies [that] lead development of the most advanced AI models and technologies, conduct insightful research into the use of frontier AI, and pioneer efforts to address both the potential benefits and risks of frontier AI technologies.”
Unanswered was whether any of the “agentic AI” will be tasked with “warfighting” directly — flying planes, shooting missiles, etc. — or if its all gonna be other stuff, like payroll or checking IDs at the gate or similiar mundane tasks that could not possibly screw up your life if an AI did it wrong.
But there are some hints. The release noted that the CDAO will also be “providing access” in the project to several existing AI systems inside the Pentagon.
Those include:
- Maven Smart System, an AI-powered targeting system the Pentagon has been building to comb through satellite and drone imagery, and other information like geolocation data, to find targets for artillery and other weapons.
- The Army Enterprise LLM Workspace, which appears to handle routine office tasks “from drafting press releases to reclassifying personnel descriptions, and everything in between,” according to an Army description of the system.
- ADVANA, the Pentagon’s massive computerized data hub, which collects data from logistics agencies, space organizations, special ops, and other areas for analysis.
The near-billion-dollar announcement is just the latest of Silicon Valley’s successful AI-powered inroads at the Pentagon. Last month, the Army direct-commissioned four senior Silicon Valley business leaders into the Reserve as lieutenant colonels. At least two of those new officers hold, or recently held, full-time, highly lucrative positions at OpenAI.
xAI belongs to Elon Musk, who famously spent the spring shutting down government computer systems across many agencies via the Department of Government Efficiency, or DOGE, and now, by coincidence, is selling an AI that just so happens to specialize in government agencies that for some reason no longer have working computers. He literally calls it “Grok for Government.”
The latest on Task & Purpose
- Air Force updates uniform standards including new rules for boots
- The Army and Navy want the ‘right to repair’ their own equipment
- Here is every rifle Marines have used in the last 250 years
- The Army has realized that horses are no longer good for ‘warfighting’
- Army will look for false accusations, consider ‘credibility’ in misconduct cases
