Now, the Pentagon is intent on fielding multiple thousands of relatively inexpensive, expendable\u00a0AI-enabled autonomous vehicles<\/a>\u00a0by 2026 to keep pace with China. The ambitious initiative \u2014 dubbed Replicator \u2014 seeks to \u201cgalvanize progress in the too-slow shift of U.S. military innovation to leverage platforms that are small, smart, cheap, and many,\u201d Deputy Secretary of Defense Kathleen Hicks said in\u00a0August<\/a>.<\/p>\n While its funding is uncertain and details vague, Replicator is expected to accelerate hard decisions on what AI tech is mature and trustworthy enough to deploy – including on weaponized systems.<\/p>\n There is little dispute among scientists, industry experts and Pentagon officials that the U.S. will within the next few years have fully autonomous lethal weapons. And though officials insist humans will always be in control, experts say advances in data-processing speed and machine-to-machine communications will inevitably relegate people to supervisory roles.<\/p>\n That\u2019s especially true if, as expected, lethal weapons are deployed en masse in\u00a0drone swarms.<\/a>\u00a0Many countries are working on them \u2014 and neither China, Russia, Iran, India or Pakistan have signed\u00a0a U.S.-initiated pledge<\/a>\u00a0to use military AI responsibly.<\/p>\n It\u2019s unclear if the Pentagon is currently formally assessing any fully autonomous lethal weapons system for deployment, as required by a 2012 directive. A Pentagon spokeswoman would not say.<\/p>\n Replicator highlights immense technological and personnel challenges for Pentagon procurement and development as the AI revolution promises to transform how wars are fought.<\/p>\n \u201cThe Department of Defense is struggling to adopt the AI developments from the last machine-learning breakthrough,\u201d said Gregory Allen, a former top Pentagon AI official now at the Center for Strategic and International Studies think tank.<\/p>\n The Pentagon\u2019s portfolio boasts more than 800 AI-related unclassified projects, much still in testing. Typically, machine-learning and neural networks are helping humans gain insights and create efficiencies.<\/p>\n \u201cThe AI that we\u2019ve got in the Department of Defense right now is heavily leveraged and augments people,\u201d said Missy Cummings, director of George Mason University\u2019s robotics center and a former Navy fighter pilot.\u201d \u201cThere\u2019s no AI running around on its own. People are using it to try to understand the fog of war better.\u201d<\/p>\n Space, war\u2019s new frontier<\/strong><\/p>\n One domain where AI-assisted tools are tracking potential threats is space, the latest frontier in military competition.<\/p>\n China envisions using AI, including on satellites, to \u201cmake decisions on who is and isn\u2019t an adversary,\u201d U.S. Space Force chief technology and innovation officer Lisa Costa, told an online conference this month.<\/p>\n The U.S. aims to keep pace.<\/p>\n An operational prototype called Machina used by Space Force keeps tabs autonomously on more than 40,000 objects in space, orchestrating thousands of data collections nightly with a global telescope network.<\/p>\n Machina\u2019s algorithms marshal telescope sensors. Computer vision and large language models tell them what objects to track. And AI choreographs drawing instantly on astrodynamics and physics datasets, Col. Wallace \u2018Rhet\u2019 Turnbull of Space Systems Command told a\u00a0conference in August<\/a>.<\/p>\n Another AI project at Space Force analyzes radar data to detect imminent adversary missile launches, he said.<\/p>\n Maintaining planes and soldiers<\/strong><\/p>\n Elsewhere, AI\u2019s predictive powers help the Air Force keep its fleet aloft, anticipating the maintenance needs of more than 2,600 aircraft including B-1 bombers and Blackhawk helicopters.<\/p>\n Machine-learning models identify possible failures dozens of hours before they happen, said Tom Siebel, CEO of Silicon Valley-based C3 AI, which has the contract. C3\u2032s tech also models the trajectories of missiles for the\u00a0the U.S. Missile Defense Agency<\/a>\u00a0and identifies insider threats in the federal workforce for the\u00a0Defense Counterintelligence and Security Agency<\/a>.<\/p>\n Among health-related efforts is a pilot project tracking the fitness of the Army\u2019s entire Third Infantry Division \u2014 more than 13,000 soldiers. Predictive modeling and AI help reduce injuries and increase performance, said Maj. Matt Visser.<\/p>\n In Ukraine, AI provided by the Pentagon and its NATO allies helps thwart Russian aggression.<\/p>\n NATO allies share intelligence from data gathered by satellites, drones and humans, some aggregated with software from U.S. contractor Palantir. Some data comes from\u00a0Maven<\/a>, the Pentagon\u2019s pathfinding AI project now mostly managed by the National Geospatial-Intelligence Agency, say officials including retired Air Force Gen.\u00a0Jack Shanahan,<\/a> the inaugural Pentagon AI director.<\/p>\n Maven began in 2017 as an effort to process video from drones in the Middle East \u2013 spurred by U.S. Special Operations forces fighting ISIS and al-Qaeda \u2014 and now aggregates and analyzes a wide array of sensor- and human-derived data.<\/p>\n AI has also helped the U.S.-created\u00a0Security Assistance Group-Ukraine<\/a>\u00a0help organize logistics for military assistance from a coalition of 40 countries, Pentagon officials say.<\/p>\n All-Domain Command and Control<\/strong><\/p>\n To survive on the battlefield these days, military units must be small, mostly invisible and move quickly because exponentially growing networks of sensors let anyone \u201csee anywhere on the globe at any moment,\u201d then-Joint Chiefs chairman Gen. Mark Milley observed in\u00a0a June speech.<\/a>\u00a0\u201cAnd what you can see, you can shoot.\u201d<\/p>\n To more quickly connect combatants, the Pentagon has prioritized the development of intertwined battle networks \u2014 called Joint All-Domain Command and Control \u2014 to automate the processing of optical, infrared, radar and other data across the armed services. But the challenge is huge and fraught with bureaucracy.<\/p>\n Christian Brose, a former Senate Armed Services Committee staff director now at the defense tech firm Anduril, is among military reform advocates who nevertheless believe they \u201cmay be winning here to a certain extent.\u201d<\/p>\n \u201cThe argument may be less about whether this is the right thing to do, and increasingly more about how do we actually do it — and on the rapid timelines required,\u201d he said. Brose\u2019s 2020 book, \u201cThe Kill Chain,\u201d argues for urgent retooling to match China in the race to develop smarter and cheaper networked weapons systems.<\/p>\n To that end, the U.S. military is hard at work on \u201chuman-machine teaming.\u201d Dozens of uncrewed air and\u00a0sea vehicles<\/a>\u00a0currently keep tabs on\u00a0Iranian activity<\/a>. U.S. Marines and Special Forces also use Anduril\u2019s autonomous\u00a0Ghost mini-copter<\/a>, sensor towers and counter-drone tech to protect American forces.<\/p>\n Industry advances in computer vision have been essential. Shield AI lets drones operate without GPS, communications or even remote pilots. It\u2019s the key to its\u00a0Nova, a quadcopter<\/a>, which U.S. special operations units have used in conflict areas to scout buildings.<\/p>\n On the horizon: The Air Force\u2019s \u201cloyal wingman\u201d program intends to pair piloted aircraft with autonomous ones. An F-16 pilot might, for instance, send out drones to scout, draw enemy fire or attack targets. Air Force leaders are aiming for a debut\u00a0later this decade<\/a>.<\/p>\n The race to full autonomy<\/strong><\/p>\n The \u201cloyal wingman\u201d timeline doesn\u2019t quite mesh with Replicator\u2019s, which many consider overly ambitious. The Pentagon\u2019s vagueness on Replicator, meantime, may partly intend to keep rivals guessing, though planners may also still be feeling their way on feature and mission goals, said Paul Scharre, a military AI expert and author of \u201cFour Battlegrounds.\u201d<\/p>\n Anduril and Shield AI, each backed by hundreds of millions in venture capital funding, are among companies vying for contracts.<\/p>\n Nathan Michael, chief technology officer at Shield AI, estimates they will have an autonomous swarm of at least three uncrewed aircraft ready in a year using its\u00a0V-BAT aerial drone<\/a>. The U.S. military currently uses the V-BAT — without an AI mind — on Navy ships, on counter-drug missions and in support of Marine Expeditionary Units, the company says.<\/p>\n It will take some time before larger swarms can be reliably fielded, Michael said. \u201cEverything is crawl, walk, run — unless you\u2019re setting yourself up for failure.\u201d<\/p>\n The only weapons systems that Shanahan, the inaugural Pentagon AI chief, currently trusts to operate autonomously are wholly defensive, like Phalanx anti-missile systems on ships. He worries less about autonomous weapons making decisions on their own than about systems that don\u2019t work as advertised or kill noncombatants or friendly forces.<\/p>\n The department\u2019s current chief digital and AI officer Craig Martell is determined not to let that happen.<\/p>\n \u201cRegardless of the autonomy of the system, there will always be a responsible agent that understands the limitations of the system, has trained well with the system, has justified confidence of when and where it\u2019s deployable — and will always take the responsibility,\u201d said Martell, who previously headed machine-learning at LinkedIn and Lyft. \u201cThat will never not be the case.\u201d<\/p>\n As to when AI will be reliable enough for lethal autonomy, Martell said it makes no sense to generalize. For example, Martell trusts his car\u2019s adaptive cruise control but not the tech that\u2019s supposed to keep it from changing lanes. \u201cAs the responsible agent, I would not deploy that except in very constrained situations,\u201d he said. \u201cNow extrapolate that to the military.\u201d<\/p>\n Martell\u2019s office is evaluating potential generative AI use cases \u2013 it has a\u00a0special task force<\/a>\u00a0for that \u2013 but focuses more on testing and evaluating AI in development.<\/p>\n One urgent challenge, says\u00a0Jane Pinelis, chief AI engineer<\/a>\u00a0at Johns Hopkins University\u2019s Applied Physics Lab and former chief of AI assurance in Martell\u2019s office, is recruiting and retaining the talent needed to test AI tech. The Pentagon can\u2019t compete on salaries. Computer science PhDs with AI-related skills can earn more than the military\u2019s top-ranking generals and admirals.<\/p>\n Testing and evaluation standards are also immature, a\u00a0recent National Academy of Sciences<\/a>\u00a0report on Air Force AI highlighted.<\/p>\n Might that mean the U.S. one day fielding under duress autonomous weapons that don\u2019t fully pass muster?<\/p>\n \u201cWe are still operating under the assumption that we have time to do this as rigorously and as diligently as possible,\u201d said Pinelis. \u201cI think if we\u2019re less than ready and it\u2019s time to take action, somebody is going to be forced to make a decision.\u201d<\/p>\n![\"\"](\"https:\/\/www.fie.undef.edu.ar\/ceptm\/wp-content\/uploads\/2023\/12\/LLBCXBYYCBCZBB6F2AISWQRMGQ-1024x683.jpg\")
![\"\"](\"https:\/\/www.fie.undef.edu.ar\/ceptm\/wp-content\/uploads\/2023\/12\/DSZPZY7XOVFTBNWHYJJW4GA4XI-1024x539.jpg\")
![\"\"](\"https:\/\/www.fie.undef.edu.ar\/ceptm\/wp-content\/uploads\/2023\/12\/Y3SKU6Z27BH7TH5LY6VNJ2RDB4.jpg\")
![\"\"](\"https:\/\/www.fie.undef.edu.ar\/ceptm\/wp-content\/uploads\/2023\/12\/ECKQJC65Z5H27OLMSUF4ODP2LM.jpg\")