AI Library
The Age of Autonomous Scientific Discovery
Kim Kyung-jin, Attorney at Law
AI Scientists and Self-Driving Labs
This book follows how AI scientists and self-driving labs are changing the way science generates and verifies claims. It covers literature-based discovery, natural-language protocols translated into robot commands, multi-agent research systems, closed-loop laboratories, materials search, the verification gap, chains of evidence, research harnesses, journal ethics, and legal responsibility.
AI Library
A New Era of Life Sciences Opened by Artificial Intelligence
Structural Proteomics, Genomic Foundation Models, Autonomous Laboratories, and Global Governance
Kim Kyung-jin, Attorney at Law
This book is a research volume compiled with artificial intelligence. A human selected the materials and structured the work, while AI models drafted the sentences and cross-checked the facts.
AI Library
The Double Structure of Digital Sovereignty
Europe’s Departure from Palantir and the Chains of American Big Tech
Kim Kyung-jin, Attorney at Law
This is a record of 2026, when European intelligence agencies and defense ministries began removing analytics tools from America’s Palantir. It covers the replacement decisions made by France’s General Directorate for Internal Security (DGSI), Germany’s Federal Office for the Protection of the Constitution (BfV), and the Netherlands Ministry of Defense; the incident in which US export controls severed an ally’s ac…
New English Edition
Artificial Intelligence in Horticulture
Kim Kyung-jin, Attorney at Law
Across five chapters and ten sections, this book examines computer vision for crop diagnosis, harvesting robots and autonomous field systems, smart greenhouses and digital twins, precision irrigation and supply-chain quality control, high-throughput phenotyping, and predictive breeding.
New English Edition
Artificial Intelligence in Food Crop Agriculture
Kim Kyung-jin, Attorney at Law
Across six chapters and eighteen sections, the book examines digital agricultural infrastructure, remote sensing, crop diagnosis, yield forecasting, precision irrigation, genomics, molecular breeding, agricultural robotics, climate-smart agriculture, and global food security.
New English Edition
The Future of Forestry and Agroforestry
Kim Kyung-jin, Attorney at Law
Driven by Artificial Intelligence and Digital Innovation
Across five chapters and fifteen sections, the book follows satellites, drones, LiDAR, digital twins, forest-specific language models, wildfire and pest forecasting, forestry robotics, agroforestry, timber traceability, and forest carbon markets.
New English Edition
Smart Livestock Farming: AI Enters the Barn
Kim Kyung-jin, Attorney at Law
Sensors listen, cameras watch, and artificial intelligence helps farmers decide.
Across five chapters and fifteen sections, the book follows precision livestock farming from animal health and reproduction to robotic milking, virtual fencing, digital twins, methane reduction, welfare, and data ownership.
Table of Contents
Han Dong-hoon, Busan Buk-gu Gap: A Record of the 100 Days Before and After the Election (Mar. 26-Jul. 3, 2026)
Kim Kyung-jin
Table of Contents and 13 sections
From March 26 to July 3, 2026, this record follows the spring after expulsion, the Busan Buk-gu Gap by-election, victory as an independent, and the first bill submitted in the National Assembly.
Table of Contents
Artificial Intelligence and Medicine
Kim Kyung-jin, Attorney at Law
AI in clinical care, hospitals, education, and research
AI in medical imaging, risk prediction, treatment planning, hospital operations, education, and research, with patient safety, privacy, and accountability.
[AI Library] Chapter 5. Ukraine: The Testing Ground for AI Warfare
PALANTIR: War, Surveillance, Artificial Intelligence
Part 3: AI on the Battlefield, Software-Defined Warfare
Chapter 5. Ukraine: The Testing Ground for AI Warfare
Attorney Kyungjin Kim
A. How David Defeats Goliath: Accelerating the Kill Chain
(1) Real-Time Integration of Satellite, Drone, and HUMINT Intelligence
On a morning in October 2022, at an undisclosed location on the outskirts of Kyiv, two Ukrainian officers sat before a laptop. On the screen, a digital map of the Bakhmut front was spread out. The distance between the Russian and Ukrainian trenches was barely a few hundred meters. When a technician clicked the mouse, a thermal image appeared. Another click. This time, a photograph of a Russian tank marked with the letter Z appeared, taken by a ground scout through a fence. Had this been an actual operation, they would have chosen a missile, artillery, or armed drone and issued an attack order within minutes.
Washington Post reporter David Ignatius, who witnessed this scene firsthand, wrote: "This could be a revolution in warfare."
What he saw was Palantir's software. The system integrates multiple eyes on a single screen: images from commercial satellites, real-time video from military drones, photos uploaded by frontline scouts via smartphones, enemy artillery positions detected by counter-battery radar, and even intelligence obtained by hacking Russian surveillance cameras. In the past, each piece of data was trapped in different departments, different systems. It was common for one agency to know something that another did not. It was the exact same problem U.S. intelligence agencies experienced during the 9/11 attacks.
Palantir CEO Alex Karp publicly stated that the company's technology is responsible for a significant portion of Ukrainian targeting intelligence. The company established an office in Kyiv and dispatched engineering teams. They are constantly experimenting and improving on the battlefield.
Modern warfare is a race against entropy. When decisions are delayed, targets move, fire, and hide. When Ukraine received precision weapons like HIMARS from the West, the success of those weapons depended less on range than on the speed and quality of data. Palantir's technology is called "software that compresses time" , because it dramatically shortens the time from target discovery to analysis to strike.
Ukraine also possesses its own domestically developed system called DELTA. Started in 2016 by a volunteer organization called Aerorozvidka, this system was officially deployed across the entire defense and security sector in August 2024. With tens of thousands of users, DELTA integrates satellite images, drone footage, radar data, cyber intelligence, and even information reported by citizens through Telegram chatbots. In July 2024, DELTA passed a NATO information security audit , the first Ukrainian military system to receive NATO protocol certification.
A 2025 CSIS report assessed that DELTA could rival the U.S. CJADC2 (Combined Joint All-Domain Command and Control) system in terms of design objectives and operational flexibility. Lieutenant Colonel Boyko, who led DELTA's development, said: "We've built modules that have never been used even in NATO."
In April 2025, reports emerged that a NATO member state had begun negotiations to adopt DELTA. If this materializes, it would be the first export of a Ukrainian combat system. Technology born on the battlefield may now be reverse-exported to the world's most powerful military alliance.
(2) The Secret Behind SAKER Drones and the 50% to 80% Improvement in Drone Strike Rate
By late 2023, Ukrainian drone pilots were in despair. Novice operators' hit rates had dropped to as low as 10%. Even veterans struggled to exceed 50%. The reason was electronic warfare. The Russian military had deployed jamming equipment in large quantities to disrupt drone signals. Drones dependent on GPS lost their bearings, and drones that lost their link to the operator crashed to the ground or flew to unintended locations.
That was when SAKER Scout appeared.
SAKER is a reconnaissance drone. It is equipped with Palantir's artificial intelligence software. SAKER identifies targets on its own , tanks, armored vehicles, troop concentrations. The AI determines what they are. It then transmits that information to the command post. The commander decides which weapon to use and when to strike.
SAKER's range is 10 kilometers. It uses inertial navigation instead of GPS. It is resistant to jamming. But the real secret lies not in the hardware but in the software.
Palantir's AI learns. Ukrainian operators fed it Russian military footage , thousands of hours of video. Russian soldiers' uniforms, equipment, weapons. The way they walk, their movement patterns. The AI memorized all of it. Now SAKER can find camouflaged equipment. If a person not wearing a uniform moves like a Russian soldier, the AI recognizes it.
In 2024, according to reports by Time magazine and Britain's Forces News, AI drone hit rates climbed to 80%. From below 50% to 80%. This means nearly twice as many enemy targets can be struck with the same number of drones.
David Kirichenko, writing for the Henry Jackson Society, analyzed: "AI-equipped drones are using technology to counter Russia's numerical advantage in manpower and weapon systems. Ukraine has squarely overturned the conventional wisdom that a much larger great power would overwhelm a smaller country."
But he also expressed concern. If AI can identify Russian soldiers not wearing uniforms based on movement patterns alone, how would it distinguish Russian civilians? Kirichenko said that Ukrainian and Western militaries would halt fire in that situation, but he could not be sure what would happen if the Russian military acquired the same technology.
Forces News concluded: "The wars of the future will be as much about algorithms as armor."
In September 2024, Ukraine's Ministry of Defense officially approved SAKER Scout for military supply. This was thanks to the Ministry of Defense accelerator program established in June. The program ensures that proven innovative technologies can be rapidly deployed to the field without bureaucratic delays. A process that would have taken two years in the past was reduced to a month and a half.
SAKER also integrates with the DELTA system. When a reconnaissance drone discovers a target, that information is immediately displayed on DELTA's digital map. The commander can see available fire assets on the same screen , which battery is within range, which armed drone has completed pre-flight preparation. Once a decision is made, FPV kamikaze drones included in the SAKER system fly toward the target.
Humans make the final decision. For now. But AI's role is growing. Finding, classifying, and recommending targets is now the machine's job. What remains for the human is pressing the approval button.
A Ukrainian Ministry of Defense official said, on condition of anonymity: "First-person-view drones with AI could achieve an 80% hit rate."
The future of warfare is flying over the trenches of Donetsk.
B. Command and Control by Algorithm
(1) Shortening the Time from Target Identification to Strike Approval
In military terminology, there is a concept called the "Kill Chain." It refers to the series of steps: finding a target, tracking it, identifying it, assigning a weapon, attacking, and assessing the result. Each link in this chain consumes time. In traditional methods, this process could take 72 hours. Assuming the target would still be at the location where it was spotted was closer to wishful thinking.
Ukraine has dramatically shortened this time. In some cases, to just two minutes.
The secret was not replacing human judgment but eliminating the time spent waiting for human judgment. In the past, when a reconnaissance unit collected intelligence, it went up to higher command, was reviewed by analysts, decided upon by a commander, and transmitted back down. Reports were written at each step. Meetings were convened. Approval stamps were applied.
DELTA and Palantir systems compress all of this into a single screen.
According to a report published in June 2025 by Ukrainian security expert Kateryna Aniskina, DELTA works as follows: ISTAR (Intelligence, Surveillance, Target Acquisition, Reconnaissance) units collect raw data from all available sources , informant reports, automated collection through chatbots, satellite images, aerial reconnaissance data, radar, cyber data, and more. Then various actors in the security and defense sectors input this information into DELTA. It layers onto the digital map.
What matters here is standardization. The system's tools convert data into a machine-readable standard format. Room for interpretation is removed. An object on the map , say, a specific weapon or piece of military equipment , is exactly that object as entered into the system. It cannot be interpreted differently.
The information processed this way is objectively displayed on the digital map in near real time. And it immediately becomes accessible to all other actors who need battlefield situational awareness.
DELTA is composed of several modules. Monitor is the situational awareness map. Target Hub manages fire missions. Vezha provides real-time video streams. Mission Control handles unmanned vehicle coordination. Element is secure chat. According to Defense Minister Umerov, the value of enemy equipment destroyed through DELTA exceeds $15 billion.
A 2025 Carnegie Endowment for International Peace report noted: "Despite advances in reconnaissance, digitization, AI, and predictive analytics, war remains kinetic, chaotic, and destructive. Technology has shaped decisions but has not eliminated the fog and friction of war."
But the fog has clearly thinned.
There is an AI system called Griselda. This system processes collected data in an average of 28 seconds. The processed information is transmitted to the DELTA battle management system and the GIS Arta fire support system. Twenty-eight seconds. In the time a human analyst reads the first paragraph of a report, the AI has already reached its conclusion.
The true meaning of algorithmic command and control lies here. It is not about improving the quality of decisions. It is about reducing the time to reach a decision. In war, time is life. Literally.
(2) Data Visualization that Clears the Fog of War
Prussian military theorist Carl von Clausewitz coined the concept of the "fog of war." In warfare, commanders must make decisions amid incomplete information, contradictory reports, and unpredictable circumstances. Where the enemy is, how many there are, what they intend to do , almost nothing is known with certainty.
Ukraine is trying to clear this fog. Even if it cannot be completely eliminated, the goal is to make it as thin as possible.
DELTA's digital map is not an ordinary map. It is a living map. When a satellite transmits a new image, the map updates. When a drone sends video, it appears on the map. When a ground scout takes a photo with a smartphone and uploads it, that too appears on the map. Everything in real time. Everything on one screen.
The Vezha module handles video streaming. A commander can simultaneously watch footage from dozens of drones. In September 2024, at NATO's REPMUS 24 exercise in Portugal, DELTA coordinated more than 50 drones , underwater drones, surface drones, ground robots, aerial drones, and even Rheinmetall's Robotics-L "robot dog." All under a single system.
The power of visualization lies in simplifying complexity. Raw data is meaningless , just a list of numbers and coordinates. But when it is displayed as icons on a map, color-coded, and changes over time are shown as animation, the human brain recognizes patterns. What is important, what is dangerous, what should be dealt with first.
Palantir's system works the same way. In the Washington Post article, the reporter saw a screen with the Bakhmut front's trenches drawn in fine detail. A click brings up thermal imagery. Another click shows a photo taken by a ground informant. All information is layered on the same map. A commander can feel as though viewing the battlefield from a god's-eye perspective.
But there are dangers too: the temptation to believe that what is on the screen is everything. What the sensors missed, what the data misinterpreted, what the AI misidentified , those do not appear on the screen. Clausewitz's fog has not completely disappeared. It has merely changed form.
A 2024 report from the German Marshall Fund warned: "The use of companies like Palantir and the DELTA system has the potential to affect democracy and human rights. The use of advanced data mining tools raises concerns about misuse and privacy violations. The fact that the company has classified contracts with governments underscores the need for ethical use and strict compliance with privacy laws."
Data visualization is a double-edged sword. It gives commanders unprecedented situational awareness. But at the same time, it can breed a new form of ignorance , overconfidence in technology. Believing only what the screen shows and thinking that what the screen does not show does not exist. That is the new fog.
C. GIS Arta and the Maven Smart System
(1) The "Uber-Style" Fire Support System
May 2022. The Siverskyi Donets River in the Luhansk region. Russian forces tried to cross the river. They set up pontoon bridges and attempted to move tanks and armored vehicles to the other side. Days later, satellite photos showed destroyed pontoon bridges and burning vehicles. In two days, Russian forces lost dozens of tanks and armored vehicles. Personnel losses were estimated between 1,000 and 1,500.
Behind this battle was a piece of software called GIS Arta.
GIS Arta , an abbreviation for Geographic Information System for Artillery. Ukrainian developers reportedly began building it around 2015. People call it "Uber for artillery."
Here is how it works. A scout on the front line spots the enemy position. The scout takes a photo with a smartphone and uploads it along with GPS coordinates. There is also footage from drones. There are also enemy artillery positions detected by counter-battery radar. All of this information flows into a central server.
The server's algorithm begins calculating. What type of target is it? Where is it located? Which battery is within range? Which ammunition is appropriate? If infantry is exposed, mortars are effective. If vehicles, howitzers are better. The algorithm finds the optimal combination.
Then it assigns the "job" , just as Uber assigns the nearest driver to a passenger.
Artillery units connect to the network and mark themselves as "available." It is the same as an Uber driver turning on the app and waiting for passengers. When a fire mission comes in, available batteries within range are highlighted. The commander touches the screen. From target acquisition to firing: 30 to 45 seconds.
Another strength of GIS Arta is dispersion. In traditional artillery operations, batteries were clustered in one location for concentrated fire. But concentration is vulnerable to counter-battery fire. If the enemy detects the firing position, return fire pours in.
GIS Arta is different. Multiple guns at different locations can simultaneously attack the same target. The system calculates when the shells fired from each gun will reach the target. It then coordinates firing times so that all shells land on the target at nearly the same moment. A choreography of shells arriving from multiple directions , making it difficult for Russian forces to respond.
Ukrainian batteries practice "shoot and scoot." They fire, then relocate within minutes. In the past, this took hours. It became possible because GIS Arta can assign new fire missions immediately even after repositioning.
On some fronts, Russian artillery outnumbers Ukraine's by a factor of 20. GIS Arta is a "force multiplier" that narrows this gap. Fewer guns strike more targets, more accurately, more quickly, with less risk.
The New America Foundation's analysis put it this way: "GIS Arta is essentially like any other app you can find on your phone or computer. Like Uber or Google Maps, it requires infrastructure and human capital to maintain."
The system requires internet connectivity. In February 2022, Russian forces cyberattacked Tooway, a satellite internet provider. Ukraine's combat systems were paralyzed. That was when Elon Musk's Starlink appeared. Thousands of terminals arrived in Ukraine. GIS Arta began working again.
The Battle of Siverskyi Donets was the result.
(2) The Reality of AI Tactics That Halted Russian Armored Offensives
On March 25, 2025, the NATO Communications and Information Agency (NCIA) signed a contract with Palantir. It was the adoption of Maven Smart System NATO (MSS NATO). From requirements definition to system acquisition in just six months , one of the fastest procurement processes in NATO's history.
Why the urgency?
NATO had been watching what was happening in Ukraine. Software was changing the nature of war. DELTA, GIS Arta, and Palantir's technology had enabled the numerically inferior Ukrainian military to stand and fight against Russian forces. NATO wanted to make these lessons their own.
The Maven Smart System was originally developed for the U.S. Department of Defense. In 2021, it was derived from Project Maven's object recognition AI. Initially, the purpose was to find terrorist targets in surveillance footage. But the system evolved. It now integrates classified and open-source materials into a single searchable database. Recently, generative AI and large language model (LLM) capabilities have been added.
According to the NATO press release, MSS NATO "provides the Alliance with a common data-driven warfighting capability. Through a range of AI applications from large language models to generative AI to machine learning, it improves intelligence fusion and targeting, battlefield awareness and planning, and decision acceleration."
In plain terms, it connects the different databases of different NATO member nations into one. U.S. intelligence, German intelligence, French intelligence all appear on the same screen. This was previously impossible. Each nation's system was incompatible. Formats differed. Security protocols differed.
General Markus Laubenthal, Chief of Staff at NATO's Supreme Headquarters Allied Powers Europe (SHAPE), said: "Innovation is key to our warfighting capability. Maven Smart System NATO enables the Alliance to harness complex data, accelerate decision-making, and thereby add genuine operational value."
Allied Command Operations (ACO) announced it would begin using the new system within 30 days of contract signing, with operations expected from May 2025. According to a SHAPE spokesperson, Maven was already "actively validated, contributing to live operations across nine NATO member states."
Contract sizes in the United States are also surging. In May 2024, the Pentagon signed a five-year contract with Palantir worth $480 million. In May 2025, the contract ceiling was increased by $795 million to nearly $1.3 billion. The reason cited was "increased demand from combatant commands."
According to Breaking Defense, MSS has already become "indispensable" to the Pentagon's Joint Staff and geographic combatant commands, including European Command. A 2024 CDAO (Chief Digital and Artificial Intelligence Office) contract expanded the user base from "hundreds" to "thousands."
According to Janes, the system is designed to link multiple command, control, and communications (C3) systems "to share and exploit data from legacy systems, third-party apps, and even bespoke solutions in ways that were previously impossible."
There are dissenting voices. An analysis by Defense Update notes: "The selection of Palantir, a U.S. technology company, as the central system for the entire Alliance highlights the ongoing tension between the need to use available cutting-edge technology (often originating from the U.S.) for immediate capability improvement and Europe's long-term strategic autonomy aspirations to reduce technological dependence on non-European suppliers."
A 2025 analysis from the Carnegie Endowment for International Peace raises a more fundamental question: "Commercial actors like Palantir are neither neutral suppliers nor patriotic agents. Their primary motivations are commercial, shaped by venture capital, shareholder expectations, and market valuations."
Analysts warn: if future commercial interests are judged to lie elsewhere, companies may shift their loyalties.
Ukraine has become the testing ground for AI warfare. The products of that experiment are now being exported to the world's most powerful military alliance. Algorithms verified on the battlefield are being transplanted into peacetime planning systems. Software is defining the future of war.
The question is who controls that software.
Kim Kyung-jin
Attorney · Former Member of the National Assembly · AI Policy Researcher
© 2026 Kim Kyung-jin. All rights reserved.