US Air Force launches $99 million AI effort to track threats before they strike
August 11, 2025
With warfare evolving rapidly from hypersonic missiles to autonomous drone swarms, the United States Air Force has unveiled a high-stakes initiative to develop cutting-edge target tracking systems powered by artificial intelligence.
Backed by a potential $99 million in funding, the programme seeks to fuse massive data streams with advanced machine learning and inferencing algorithms, housed within high-performance computing environments.
The ambition is simple but audacious: to transform traditional surveillance into predictive, machine-driven warfare.
The new effort is being spearheaded by the Air Force Research Laboratory (AFRL), operating out of its Information Directorate in Rome, New York.
Formally titled “Advanced Tracking Architecture Using AI” (ATA-AI), the research call invites white papers and proposals from scientists, engineers and defence contractors capable of building the next generation of target tracking systems.
The open call, technically known as an Advanced Research Announcement (ARA), is open through 12 July 2029 and reflects a fundamental shift in how the US military hopes to outpace rapidly advancing adversaries.
Using AI to track threats before they move
Target tracking has long been a cornerstone of air and missile defence, but in today’s battlespace saturated with stealthy aircraft, hypersonic glide vehicles and low-flying drones, traditional methods struggle to keep up.
The new initiative aims to fix that by leveraging the combined power of Artificial Intelligence (AI), Machine Learning (ML), and Machine Inferencing (MI) across a wide array of sensors and data streams.
The end goal is a system that can not only follow fast-moving targets but also predict their behaviour, even amid cluttered, contested environments.
“Tomorrow’s battles will be won not just by seeing faster, but by knowing sooner,” said a senior Indian defence official familiar with a similar project. “ATA-AI is about giving machines the ability to understand threats before they act and doing so faster than any human ever could.”
The technologies developed under this programme are expected to integrate seamlessly into command, control, communications and battle management (C3BM) systems, which are at the heart of modern military operations.
One such programme focused on moving target indication (MTI) is already being fast-tracked under a rapid acquisition authority granted by the US Congress in the fiscal year 2024 defence budget. That effort aims to field early capabilities by 2027.
Collaborative working for faster innovation
The AFRL’s open call is unique not just for its technical ambition but also for its collaborative structure. The laboratory is seeking a diverse range of participants from large defence contractors to academic research centres and tech startups, and is encouraging innovative approaches that blend agility with scientific rigour.
Initial submissions must come in the form of white papers, not full proposals, with formal invitations extended to selected offerors. Projects may be awarded using traditional procurement contracts or Other Transaction (OT) authorities, which offer greater flexibility for prototype development and commercial tech integration.
While individual awards are typically capped between $1 million and $10 million, the AFRL has left the door open to larger investments for groundbreaking concepts that push the envelope.
Multiple awards are anticipated, and each project may run for up to 48 months, depending on complexity and funding availability.
Notably, projects that demonstrate success during the prototype phase could be transitioned into full-scale production contracts, significantly accelerating the time it takes to get new capabilities into the field.
Advanced AI tracking can aid survival in the modern battlefield
The urgency of the ATA-AI initiative is no accident. In recent years, peer adversaries, particularly China and Russia, have developed increasingly sophisticated air and missile technologies designed to evade detection.
The proliferation of small, unmanned systems, often operating in swarms or at low altitude, has further complicated defensive operations. By the time a human operator can assess the threat, it may be too late.
Advanced AI-powered tracking architectures promise to change that. By analysing multiple data streams, including radar, infrared, electro-optical, satellite feeds, and even open-source intelligence, these systems can identify patterns and anomalies invisible to the naked eye.
When combined with high-performance computing, the result is a potential leap in decision-making speed and accuracy.
