Germany turns to AI for combat decision support
Germany plans to use artificial intelligence for combat decisions. Critics point out that LLMs are unreliable, Ukrainian data is biased, and the h
English · Original discussion in Spanish · Published
Germany turns to AI for combat decisions: the algorithmic delusion
Germany, that military power which has lost two world wars and recently demonstrated an operational capacity more befitting a ministry of bureaucracy than an army, has decided that artificial intelligence will help it make decisions on the battlefield. Lieutenant General Christian Freuding announced it with the solemnity of someone who believes they have found the philosopher's stone: feeding AI data from the Ukrainian conflict so that it can "deduce how the enemy has acted" and recommend countermeasures. The idea sounds modern, but it is a perfect conceptual trap.
The underlying problem: AI doesn't think, it assembles text
The basis of the project is a category error. Today's artificial intelligence, especially large language models (LLMs), are machines for predicting the next word. They don't understand context, they have no intentionality, and when forced to reason about complex situations, they hallucinate. An expert user can detect the limits; but the paradox is that the more expertise the user has, the faster the AI is forced to invent answers. Asking an LLM to analyze military tactics is like asking a parrot to command a naval battle. The result is a system that will generate plausible but groundless recommendations, and in the heat of combat it can lead to disaster.
The bias of Ukrainian data
The four years of Ukrainian data they intend to use are not neutral. They are, at best, a partial and biased snapshot; at worst, propaganda injected into the system. Garbage in, gospel out. If the AI learns from a war where information is contaminated by political interests and deception operations, its recommendations will be as reliable as Russian roulette. The mistake is believing that war is a data optimization problem. It is not: it is a chaotic, human and irrational phenomenon, where intuition and the sarracena factor weigh more than any statistic.
Germany and its tradition of bad wartime decisions
Germany's history on the battlefield is littered with resounding failures, from the two world wars to recent interventions. It won't be for lack of trying, but they have always ended up on the loser's side. Now they intend to delegate strategy to an algorithm. Nobody misses the irony. The only time Germany performed well militarily was fighting for the Spanish crown, and that was centuries ago. Investing in an AI system for combat is the perfect recipe for a disaster with technological pedigree.
The cost and the usual suspects
Projects like this lead nowhere, but they cost an arm and a leg and usually end up with Israeli companies. The suspicion that the whole thing is a racket, like cities on the moon, runs through the debate. AI as a concept has been hyped to nausea, and applying it to war is the final nail in the coffin of technological seriousness. The Algorithmic Reich that looms ahead will have no clear ideology, just KPIs for territorial control and repression of dissent based on "risk patterns". No Führer, but with a board of technocrats executing what the data dictates.
The question nobody answers is: who takes responsibility when the AI gets it wrong? Because it will get it wrong, and the consequences will be human lives. But in the German tech bubble, that seems like a minor detail.
Related discussions in the forum
Germany, that military power which has lost two world wars and recently demonstrated an operational capacity more befitting a ministry of bureaucracy than an army, has decided that artificial intelligence will help it make decisions on the battlefield. Lieutenant General Christian Freuding announced it with the solemnity of someone who believes they have found the philosopher's stone: feeding AI data from the Ukrainian conflict so that it can "deduce how the enemy has acted" and recommend countermeasures. The idea sounds modern, but it is a perfect conceptual trap.
The underlying problem: AI doesn't think, it assembles text
The basis of the project is a category error. Today's artificial intelligence, especially large language models (LLMs), are machines for predicting the next word. They don't understand context, they have no intentionality, and when forced to reason about complex situations, they hallucinate. An expert user can detect the limits; but the paradox is that the more expertise the user has, the faster the AI is forced to invent answers. Asking an LLM to analyze military tactics is like asking a parrot to command a naval battle. The result is a system that will generate plausible but groundless recommendations, and in the heat of combat it can lead to disaster.
The bias of Ukrainian data
The four years of Ukrainian data they intend to use are not neutral. They are, at best, a partial and biased snapshot; at worst, propaganda injected into the system. Garbage in, gospel out. If the AI learns from a war where information is contaminated by political interests and deception operations, its recommendations will be as reliable as Russian roulette. The mistake is believing that war is a data optimization problem. It is not: it is a chaotic, human and irrational phenomenon, where intuition and the sarracena factor weigh more than any statistic.
Germany and its tradition of bad wartime decisions
Germany's history on the battlefield is littered with resounding failures, from the two world wars to recent interventions. It won't be for lack of trying, but they have always ended up on the loser's side. Now they intend to delegate strategy to an algorithm. Nobody misses the irony. The only time Germany performed well militarily was fighting for the Spanish crown, and that was centuries ago. Investing in an AI system for combat is the perfect recipe for a disaster with technological pedigree.
The cost and the usual suspects
Projects like this lead nowhere, but they cost an arm and a leg and usually end up with Israeli companies. The suspicion that the whole thing is a racket, like cities on the moon, runs through the debate. AI as a concept has been hyped to nausea, and applying it to war is the final nail in the coffin of technological seriousness. The Algorithmic Reich that looms ahead will have no clear ideology, just KPIs for territorial control and repression of dissent based on "risk patterns". No Führer, but with a board of technocrats executing what the data dictates.
The question nobody answers is: who takes responsibility when the AI gets it wrong? Because it will get it wrong, and the consequences will be human lives. But in the German tech bubble, that seems like a minor detail.
Related discussions in the forum
- AMERICAN MEDIA ask TRUMP for the EXTRADITION OF ZAPATERO · 120 replies
- Game Over PSOE · 29 replies
- Where are the lobotomized ones from "This is normal, it has always been hot in May" · 87 replies
- The red thread. reaccionariopobres stay away. · 12873 replies
Also available in: Deutsch
Summary of a discussion on Burbuja.info - Foro de economía, actualidad y política., translated from Spanish and reviewed before publication. Read the full discussion (19 replies).
More summaries
-
Spain rejects EU relocation of unaccompanied minors
Canary Islands' costs for unaccompanied minors hit €165 million. Spain refuses EU offers to relocate them, citing political logic amid growing scrutiny.
-
Spain Absorbs 28% of EU Immigration
Spain absorbed 28% of EU migration and 65% of its population growth. The demographic engine supporting the GDP raises questions about its cost...
-
PP ramps up confrontational tweets on social media
Spain's PP increases aggressive online posts, sparking debate over potential electoral costs and digital strategy.
-
Only the far right sees value in heterosexual relationships, Spanish sociologist claims
Andrea García-Santesmases argues that feminism has sidelined heterosexual desire, leaving only the far right to view it positively. The key points of the controversy and its implications.
-
Spain: a golden age or a tin-plate one?
Debt at 120% of GDP, job insecurity and statistical spin: the figures that dismantle the narrative of Spain's new Golden Age.
-
Pensions: the cost of rejecting early-retirement cuts is just the tip of the iceberg
Refusing to scrap reduction coefficients for workers with over 40 years of contributions would cost 3,358 million euros. Spain is the European country with the most unfunded pension liabilities.