How does US plan to take on China? By building a drone fleet - but how well is it going?

The Pentagon is betting heavily on autonomous maritime drones to counter China’s growing naval power in the Pacific. Inspired by Ukraine’s successful use of cheap, kamikaze-style sea drones against Russia’s Black Sea Fleet, US defense planners envision swarms of high-tech, AI-driven vessels patrolling the Taiwan Strait and deterring a Chinese advance.

But a string of recent mishaps shows just how steep the learning curve may be.

Setbacks at sea

Last month, during a high-profile US naval test off the California coast, one autonomous boat abruptly stalled due to a software glitch, Reuters reported. Before operators could respond, another drone vessel smashed into it, vaulted over its deck, and crashed back into the water, the report said.

The vessels, built by rival defense tech firms Saronic and BlackSea Technologies, are part of the Navy’s ambitious push to field an autonomous fleet.

But this wasn’t an isolated failure. Just weeks earlier, another BlackSea vessel unexpectedly accelerated while being towed, capsizing a support boat and throwing its captain overboard.

Both incidents, according to people familiar with the program as cited by Reuters, were caused by a mix of software breakdowns and human error.

Billions at stake

The navy is investing heavily in these systems. BlackSea has received at least $160 million in commitments and is now producing dozens of drone boats each month. Saronic, a Silicon Valley startup recently valued at $4 billion, has attracted major venture backing and Pentagon prototype deals.

The effort is part of the $1 billion Replicator program launched in 2023 to rapidly acquire thousands of drones, aerial and maritime, across the military. US President Donald Trump has doubled down, with his latest defense bill allocating nearly $5 billion specifically for autonomous naval systems.

The Pentagon hopes that unlike Ukraine’s cheap, remote-controlled drones, America’s fleets of fully autonomous vessels, each costing several million dollars, can operate without direct human command, swarming together to scout, jam, and strike targets.

Can the navy adapt?

Experts, cited by Reuters, say that these early stumbles should be expected. Bryan Clark of the Hudson Institute noted the Navy would need to adapt "tactics as it better understands what the systems can do and what they can’t do."

But time is short. China is expanding its naval forces faster than the US, and Beijing has already demonstrated its own autonomous drone and submarine technologies. For Washington, the race to field a reliable drone fleet isn’t just about innovation, it’s about deterrence.

For now, America’s plan to outpace China at sea rests on drones that, at least in testing, still sometimes crash into each other.