The U.S. Air Force has selected contractors for its Collaborative Combat Aircraft program, advancing AI-enabled unmanned fighter jets designed to operate alongside human pilots while raising questions about cost, control and battlefield decision-making.
By yourNEWS Media Newsroom
The U.S. Air Force is moving ahead with a new generation of AI-enabled unmanned fighter jets, awarding contracts under its Collaborative Combat Aircraft program as the United States accelerates development of autonomous military systems amid intensifying competition with China.
The Air Force announced the contract selections Wednesday in a press release, describing the program as part of the service’s push to develop new aircraft, software and mission systems for future air superiority.
The program centers on semi-autonomous aircraft designed to fly with and support crewed fighters. Defense officials and industry leaders say the aircraft could expand combat options, reduce risk to pilots and lower long-term operating costs. Critics and human rights advocates warn that the growing use of artificial intelligence in combat systems raises serious ethical concerns, particularly if machines are allowed to select or engage targets without meaningful human control.
One of the aircraft in the program is the FQ-42, built by General Atomics. A company spokesperson told the Daily Caller News Foundation that General Atomics is well positioned for the project because of its long history building unmanned aircraft.
“Twenty-five years ago, General Atomics changed combat forever by pioneering missiles launched from MQ-1 Predators,” Mark Brinkley, a General Atomics spokesman, told the outlet. “No other nation had that capability. Almost 10 million flight hours later, we’re changing combat again with CCA [Collaborative Combat Aircraft]. Unmanned aircraft is all we do, and we do it extremely well.”
Brinkley said General Atomics had already begun building aircraft before receiving the contract award.
“For CCA, we were already building aircraft ahead of contract award. Those jets are in the pipeline, suppliers have contracts, and work is underway already,” Brinkley said. “We build a lot of things at risk, ahead of contracts, to save time for our customers and keep programs on track. We don’t have a scaling issue that you might associate with a new program or a startup company, because we’ve been doing this a long time. We have capacity and we forecast really well. So by having General Atomics in the mix, America gets a stable, nontraditional defense business but also gets real experience.”
The Air Force did not provide comment in time for publication, according to the report.
The program has drawn support from defense analysts who say unmanned combat aircraft can help commanders expand their options in contested environments.
“CCA can help commanders generate more options to sense and engage threats,” Timothy Walton, a senior fellow at the Hudson Institute, told the DCNF. “Additionally, they should have significantly lower operations and support costs compared to crewed fighters.”
The FQ-42 may use software from Collins Aerospace, demonstrating how multiple defense companies are contributing separate parts of the Collaborative Combat Aircraft ecosystem.
“Collins Aerospace, an RTX business, has been awarded a production contract by the U.S. Air Force to supply our Sidekick Collaborative Mission Autonomy software for the CCA Increment 1 program,” an RTX Corp. spokesperson told the DCNF. “We’re supporting integration and testing on General Atomics’ FQ-42A platform today.”
Walton said separating aircraft hardware from mission autonomy software could make the program more flexible.
“Separating mission autonomy from aircraft hardware not only helps prevent contractor lock-in, but also speeds development and iteration of both mission autonomy software and aircraft hardware,” Walton said.
Anduril, an AI-focused defense company, also received a contract to deliver an initial set of FQ-44 semi-autonomous fighter aircraft. The company said in a blog post that its aircraft and production process were designed with large-scale manufacturing in mind.
“A handful of aircraft will not move the needle in a great power conflict. From the design of the aircraft itself to the production system that will deliver it, we have maintained a relentless focus on the need to deliver at scale,” Anduril Vice President Mark Shushnar wrote. “The production line at Arsenal-1 is already on this path: work there is active today, and the production line itself is capable of delivering up to 150 aircraft per year in its current configuration … [Anduril can] scale it further to meet additional surges in demand.”
Anduril was also selected to help develop mission autonomy for the Air Force. In this context, mission autonomy refers to the ability of aircraft to continue operating with limited human direction if communications are disrupted by electronic warfare.
“Anduril has worked toward sophisticated Mission Autonomy from the early days of the company,” Vice President of Engineering Vik Pattabi wrote in a separate blog post. “Our experience on these programs created the foundational understanding required for us to take on a particularly advanced, complex, and high-stakes mission set: enabling a single human operator to interact and fight with a team of Collaborative Combat Aircraft (CCA).”
Shield AI is also part of the Air Force’s expanding AI combat ecosystem. The company said its Hivemind software is designed to perform functions normally handled by a pilot or operator.
“Hivemind is Shield AI’s … software that assumes the role of a human pilot or operator, enabling unmanned systems to sense, decide, and act,” Shield AI said in a statement to the DCNF. “Unlike traditional autopilots that simply follow preplanned routes, Hivemind can reroute around or engage dynamic obstacles, execute collaborative tactics with peer systems and piloted aircraft, respond to unexpected conditions, and complete missions safely and effectively as part of a human-machine team.”
Defense experts say the value of the program lies in combining the strengths of manned and unmanned aircraft.
“The advantage is that crewed and uncrewed aircraft each have respective strengths and weaknesses,” Douglas Birkey, executive director of the Mitchell Institute for Aerospace Studies, told the DCNF. “Combining them allows for the strengths of one to cover for the weaknesses of the other. We saw a basic form of this during Operation Epic Fury, when uninhabited MQ-9s were put into threat zones where we did not want to put crewed aircraft at risk.”
Birkey also said human judgment should remain part of any decision to use lethal force.
“Humans should be in the loop when use of kinetic force is in play. AQI can really help process mass quantities of data, help see patterns,” Birkey said. “But the ultimate decision to employ a missile or shoot should involve a trained person—their judgment is crucial.”
Human rights groups have long warned against fully autonomous weapons. Human Rights Watch said in 2020 that “Weapons systems that select and engage targets without meaningful human control are unacceptable and need to be prevented.”
U.N. Secretary-General António Guterres has also called for restrictions on lethal autonomous weapons. During remarks to the U.N. Security Council on artificial intelligence in 2023, he said, “I agree with the large number of experts that have recommended the prohibition of lethal autonomous weapons without human control.”
Human Rights Watch also asked British computer scientist Stuart Russell to support a ban on AI weapons in 2022.
The ethical debate has intensified as AI systems move from analysis and support roles toward greater battlefield autonomy. Supporters argue that autonomous systems can protect pilots, expand military reach and process large amounts of information faster than humans. Opponents argue that the use of lethal force must remain under direct human control and that algorithmic errors could have severe consequences.
Those concerns have been sharpened by recent questions surrounding a U.S. strike during the opening phase of the Iran War. The New York Times reported that a U.S. military strike hit an Iranian girls’ school, killing at least 175 people, many of them schoolgirls. A final investigation has not been released, and it remains unclear whether AI was involved. A preliminary investigation found that “outdated” data led to the school being mistakenly targeted.
Conservative commentator Tucker Carlson questioned whether artificial intelligence played any role in the mistake during a March 9 episode of “The Tucker Carlson Show.”
“Was there an AI program that targeted this school that thought it was something else?” Carlson asked.
The Collaborative Combat Aircraft program marks a major step in the Air Force’s effort to integrate unmanned aircraft and artificial intelligence into future combat operations. The contracts show that the service is moving from experimentation toward production, with companies emphasizing both autonomy and scale.
The program also leaves unresolved the central question surrounding AI weapons: how much control should remain with human operators as machines become increasingly capable of sensing, deciding and acting in combat.
Original article
Comments