Vice President JD Vance told graduating Air Force cadets Thursday that artificial intelligence should enhance human judgment in warfare, never replace it. Speaking at the U.S. Air Force Academy in Colorado Springs, Vance stressed that life-and-death military decisions must remain firmly in human hands as AI reshapes the battlefield.
"Use technology to make you better, but never submit to it," Vance said in his commencement address to roughly 900 graduating officers. "You are the masters of warfare. And both your minds, but also your hearts, are the opposite of artificial."
The message reflected a subtle shift in tone for an administration official who previously championed minimal AI regulation. Vance spent years in venture capital promoting the technology's potential and argued against excessive oversight. But as public anxiety about AI mounts, he has begun publicly grappling with its darker possibilities, especially in military applications.
The vice president cited Pope Leo XIV's recent theological document warning against unchecked AI advancement and praised the church's moral leadership on the issue. In an interview with NBC News, Vance acknowledged what voters fear most: how AI will reshape labor markets, distribute resources, and fundamentally alter human interaction.
"But the thing I worry about most with AI is how it will change warfare," he said.
Vance noted that American military strength has always rested on fighting with moral clarity. If that tradition continues into an AI-dominated future, he argued, humans cannot cede decision-making authority to machines.
"If the warfare of the future is to live up to the moral values of our ancestors, decisions over life and death must be made by humans and not machines," Vance told the cadets.
His remarks came as tensions simmer within the Trump administration over how to handle advanced AI systems. Defense Secretary Pete Hegseth has clashed with AI company Anthropic over military guardrails, with the Pentagon pushing to lift restrictions on how the company's products can be deployed for lethal autonomous weapons.
The internal debate has grown more urgent in recent weeks. In early April, Vance and Treasury Secretary Scott Bessent convened leading tech CEOs to discuss cybersecurity risks posed by the latest AI models. Days later, Anthropic unveiled its Mythos Preview model, which uncovered thousands of critical vulnerabilities in widely-used software.
That discovery triggered a scramble at the White House to craft a coherent AI policy. Chief of Staff Susie Wiles, National Cyber Director Sean Cairncross, and others intensified their involvement in deliberations. Competing camps have emerged: Bessent and allies favor closer scrutiny of powerful systems, while figures like former AI czar David Sacks push for lighter-touch governance.
An executive order responding to those cybersecurity threats reached Trump's desk but was not signed at the last moment. The measure would have established a voluntary testing mechanism allowing the U.S. government to examine the latest models from leading AI companies.
At a recent White House news conference, Vance stressed the administration's commitment to partnering with tech firms "to make sure that the American people are as safe as possible." He also nodded to upcoming policy moves, though specifics remain fluid as different factions haggle over priorities.
The vice president joked about his immunity from the boos that have greeted other commencement speakers touching on AI risks. "Now, you can't boo me. I'm the vice president of the United States," he said. But the quip masked a genuine tension: Vance has spent years betting on AI's promise while now wrestling publicly with its peril.
Author Sarah Mitchell: "Vance is trying to thread an impossible needle, cheering AI's commercial potential while sounding alarms about handing weapons decisions to machines, but the internal White House chaos over policy suggests nobody has figured out how to actually govern this technology."
Comments