The US military is assessing how human-machine collaboration stacks up against human-only decisions in terms of speed, accuracy, and overall effectiveness.
The US Air Force has completed its second Decision Advantage Sprint for Human-Machine Teaming (DASH 2) in Las Vegas, testing how AI can support commanders in complex battlespaces.
Early results showed AI systems generating thousands of potential solutions within seconds, about 30 times faster than human-only teams.
In just one hour, two vendors produced over 6,000 options across 20 scenarios, with accuracy levels comparable to human judgment.
Officials emphasized that the technology allows commanders to evaluate multiple courses of action simultaneously while keeping operators firmly in control.
Developers noted that small algorithm tweaks could further boost performance.
The exercise was led by the Advanced Battle Management System (ABMS) Cross-Functional Team, with support from the Air Force Research Lab, Integrated Capabilities Command, and the 805th Combat Training Squadron.
Seven teams participated, including industry partners and air force innovation groups.
“This level of output gives commanders options to execute multiple kill chains simultaneously and we’re excited about our next experiment to generate the courses of action with the machines to help illuminate risk, opportunity gain/loss, material gain/loss, among others,” said Col. John Ohlund, ABMS CFT director.
Human-Machine Teaming
DASH 2 is part of a broader US Air Force push to modernize command and control through human-machine teaming.
The initiative aims to accelerate decision-making, ensure better coordination across the joint force, and support the Pentagon’s Combined Joint All-Domain Command and Control vision.
Rather than replacing operators, the focus is on building AI tools that enhance human judgment under pressure.
Industry and air force teams co-develop and test software in operationally realistic settings, ensuring future systems are both practical and scalable.
Findings from DASH 2 suggest AI can significantly reduce workload, expand the options available to commanders, and improve the quality of battlefield decisions.
