By Kris Osborn, President, Center for Military Modernization
The Pentagon has been experimenting with unmanned fighter jets for many years now, and the technology is reasonably well evolved. The ability for autonomous flight and independent maneuvers in response to surrounding conditions is increasingly possible, and AI-enabled systems have shown great dogfighting prowess in simulations against manned fighters.
The technology has evolved to the point such that the Pentagon is preparing its new 6th-gen Next Generation Air Dominance stealth fighter and its B-21 Raider stealth aircraft for unmanned missions. Years ago, former Navy Secretary Ray Mabus said the F-35 will likely be the last “manned” fighter jet, a clear indication of the sophistication and effectiveness of autonomous technology.
This considered, the prevailing consensus among top weapons developers is that human decision-making and ultimate leadership is just as if not more critical than ever. Scientists, senior leaders, innovators and top technologist do seem consistent in that they suggest the optimal approach is to “combine” the best of both high-speed, data-analyzing AI and human cognition.
NGAD: Manned and Unmanned
Can unmanned systems, empowered by AI-generated computing, algorithms enabling autonomous targeting, maneuvering and flight path adjustments, and breakthrough kinds of data analysis, outperform any manned platforms?
The answer may, at least in some respects, be a clear yes, yet does that mean human consciousness and decision-making abilities are or will become obsolete and worthy of being replaced by AI-enabled algorithms?
Why can’t drones simply perform the entire range of missions intended for the Air Force’s and Navy’s 6th-generation manned platform? After all, the Air Force has already successfully flown unmanned fighter jets and even manned jets with an AI-enabled computerized co-pilot.
Manned-Unmanned Teaming
The reason there will still be a need for a manned platform, despite the rapid progress of AI-enabled computing, decision-making, and autonomy, is that there are attributes unique to human cognition that computers are simply not able to replicate. Certainly, AI-enabled high-speed computing can perform limitless amounts of procedural functions, analysis, and determinations without needing human intervention.
This is not only true at the moment but becoming increasingly true as technology continues to advance. However, there are specific elements of human cognition and dynamic decision-making, such as emotion, intuition, and other more subjective phenomena that computers simply cannot emulate.
Therefore, human decision-making will remain critical for the foreseeable future. It is one reason why the prevailing consensus among weapons developers is that the optimal approach needs to involve a combination of both human decision-making faculties and AI-empowered computer analysis. Both are needed, and the combination of the two is seen as the optimal approach to crafting future warfare tactics.
There is also an aerial command and control element, meaning there will be a need for a manned aircraft to function in a critical, time-sensitive command and control capacity from the air. The Air Force is already building what it calls Collaborative Combat Aircraft, or CCAs, essentially drones to fly in support of a manned NGAD 6th-generation fighter operating in a command and control capacity.
Technology is now enabling manned fighter jets to control groups of drones from the cockpit, a scenario that reduces latency and increases operational flexibility.
Operating as a “family of systems,” the CCAs will support a manned NGAD 6th-gen fighter to conduct forward surveillance, test enemy air defenses, network with ground forces, and even deliver weapons strikes when directed by a human.
Air Force Flies U-2 With AI Copilot
For the first time in history, an AI-enabled computer algorithm operated on board a military aircraft while in flight, coordinating navigational details, sensor information and reconnaissance missions alongside a human pilot.
The AI algorithm, called ARTUu, flew along with a human pilot on a U-2 Dragon Lady spy plane, performing tasks that would “otherwise be done by a pilot,” an Air Force report explained.
It is described as manned-unmanned teaming, or human-machine interface, a process intended to optimize the best of how computers and humans can perform. The human-computer team flew a reconnaissance mission during a simulated missile strike.
“ARTUu’s primary responsibility was finding enemy launchers while the pilot was on the lookout for threatening aircraft, both sharing the U-2’s radar,” the Air Force report said.
The combination represents the cutting edge or prevailing thinking regarding how best to leverage the benefits and promises of AI; given rapid advances in processing speed and AI-capable algorithms able to perform real-time analytics on fast arriving volumes of new information, computers are of course much faster and more efficient, in an exponential way, when it comes to performing crucial procedural functions.
“Putting AI safely in command of a U.S. military system for the first time ushers in a new age of human-machine teaming and algorithmic competition. Failing to realize AI’s full potential will mean ceding decision advantage to our adversaries.” Air Force Acquisition Executive William Roper said in the Air Force report.
Multiple data streams can be simultaneously gathered, pooled and organized in milliseconds, bounced off a known database of seemingly limitless volume to make comparisons, perform analyses and solve problems. Data compilation, organization and analysis, coupled with increasingly unprecedented computer processing speeds, can enable AI-empowered technology to discern critical distinctions and similarities between otherwise completely separated sources of incoming information, offering humans a previously impossible integrated picture.
At the same time, human cognition remains uniquely positioned to address certain nuances less calculable by mathematically-oriented computer programs. Many kinds of fast-changing dynamics, particularly those involving concepts or more subjectively determined variables, are at least at the moment best left to humans. For instance, to what extent could a computer discern “feelings” o
r “intent” as they may pertain to a combat engagement. Perhaps a pilot gets scared and makes an abrupt, unanticipated move? Perhaps a number of conflicting or different circumstances merge together into what might seem like a confusing mess? Can even the best AI-capable algorithms make sense of all of these kinds of phenomena? Particularly as they collide with one another? Many of these kinds of more subjective challenges, according to the best expert assessments of progress with AI, are still much better off addressed by humans.
Therefore, given that AI-empowered computing and subjective human cognition seems to possess distinct, discernable advantages, the best tactical approach to combat may be simply to leverage the best of each and combine them together. That is exactly what the Air Force is seeking to accomplish, an optimal blend of man-machine characteristics collectively able to maximize performance capability.
Kris Osborn is the Military Affairs Editor of 19FortyFive and President of Warrior Maven – Center for Military Modernization. Osborn previously served at the Pentagon as a Highly Qualified Expert with the Office of the Assistant Secretary of the Army—Acquisition, Logistics & Technology. Osborn has also worked as an anchor and on-air military specialist at national TV networks. He has appeared as a guest military expert on Fox News, MSNBC, The Military Channel, and The History Channel. He also has a Masters Degree in Comparative Literature from Columbia University.
