A radio transmission between several Russian soldiers in Ukraine in early March, captured from an unencrypted channel, shows panicked and confused comrades retreating after being shelled.
“Vostok, I’m Sneg 02. On the highway we have to turn left, damn it,” one of the soldiers said in Russian, using a codename that means “East” and “Snow 02.”
“Understood. No need to move further. Switch to defense. Done,” another replied.
The third soldier then tried to contact another person with the codename “Male 95”: “Yug 95, do you have contact with a senior? Warn him of the artillery barrage on the highway. Fireworks explode on the highway. Don’t follow the column. Move carefully. ”
The third Russian soldier continued, becoming increasingly agitated: “Go to the radio. Let me know your situation and the artillery position, roughly the type of weapon they are firing. ” Then the third soldier said again: “Name your square. Yug 95, answer my question. Name your square! ”
As the soldiers spoke, an AI was listening. Their words are automatically recorded, transcribed, translated and analyzed using a number of artificial intelligence Algorithm developed by Lining, a US company that provides AI services to intelligence analysts. While it is not clear whether the Ukrainian military intercepted communications, the use of AI systems to survey the Russian military on a large scale demonstrates the growing importance of sophisticated open source intelligence. vi in military conflicts.
Some Russian unsecured transmissions have been post online, translatedand analyzed on social networks. Other data sources, including smartphone video clips and social media posts, have also been scrutinized. But it is the use of natural language processing technology to analyze Russian military communications that is unique. For the Ukrainian military, learning about intercepted communications often involves human analysts working in a room somewhere, translating messages and interpreting commands.
The tool developed by Primer also shows the value of machine learning can become for intelligence analysis. The past decade has seen significant advances in Abilities of AI around image recognition, voice transcription, language translation and processing thanks to large neural network algorithms that learn from vast streams of training data. Available code and APIs that use AI can now transcribe voices, recognize faces, and perform other tasks, often with great accuracy. Given Russia’s advantage in numbers and artillery, intercepting communications could make all the difference for Ukrainian troops on the ground.
Primer sold AI algorithms trained to transcribe and translate phone calls, as well as algorithms that could pull out key terms or phrases. Sean Gourley, CEO of Primer, says the company’s engineers have modified these tools to perform four new tasks: Collect audio captured from web feeds that broadcast information communications are recorded using radio receiver hardware emulation software; to remove noise, including background chatter and music; to transcribe and translate Russian speech; and to highlight key statements relevant to the battlefield situation. In some cases, this involves retraining machine learning models to recognize colloquial terms for military vehicles or weapons.
Gourley says the ability to train and retrain AI models will become a key advantage in future wars. He said the company has made the tool available to outside parties but declined to say who. “We’re not going to say who’s using it or what they’re using it for,” Gourley said. Some other US companies have provide technology, information and expertise to Ukraine as it fought against Russian invaders.
The fact that some Russian militaries are using unreliable radio channels has surprised military analysts. It seems to indicate an under-resourced and well-prepared operation, Peter W. Singera senior member at think tank New America modern war expert. “Russia has used open means of communication to target its adversaries in previous conflicts like Chechnya, so they, of all forces, should have known the risks.” , Singer said. He added that these signals could certainly have helped the Ukrainians, although the analysis was most likely done manually. “It’s indicative of equipment failures, some arrogance, and possibly, a higher level of desperation in the Russian military,” it added. Mick Ryana retired Australian general and author.
