When the Ukrainians disassembled the Russian weapons, they found them full of Western electronics, mostly American.
This applies to Russia's highly successful Lancet kamikaze drone, also known as a loitering munition. In the Lancet-3, a key component of its alleged artificial intelligence (AI) capability is an advanced development module called the Jetson TX2, developed by American chipmaker Nvidia.
Nvidia describes the Jetson TX-2 as “the fastest, most powerful embedded AI computing device ever. This 7.5-watt supercomputer on a module brings true AI computing to the edge. It's NVIDIA Pascal™ -The family is built around the GPU and is packed with 8GB of memory and 59.7GB/s of memory bandwidth, featuring a variety of standard hardware interfaces that make it easy to integrate into a wide range of products and form factors. .
(GPU stands for Graphical Processing Unit. NVIDIA Pascal is a microarchitecture implemented on its GPUs.)
Nvidia has already moved on to a new module called Jetson Xavier NX, which is a much more powerful AI module. Nevertheless, the Jetson TX-2 is available and will be around until 2028, according to Nvidia.
Nvidia's AI modules are based on highly advanced manufacturing techniques. The main AI integrated circuit is made in Taiwan but the entire Jetson TX-2 module is assembled at BYD Huizhou in China, with another source at Foxconn in Taoyuan, Taiwan.
The module consists of several integrated circuits and looks like this:
Some of the other components of the TX-2 module come from different places, including China and South Korea. Other American and European products appear in the Lancet series and Iran's drones.
A particularly important component is the U-Blox Lea-m8s-0-10 GPS navigation system. The device can receive navigation signals from the US GPS system, Europe's Galileo GPS, Russia's Glonass and China's Beidou.
Many cell phones can do this as well (though usually not Galileo). What makes the U-Blox special, according to experts with experience dealing with the Lancet, is that it is jam and spoof resistant, meaning the GPS lock that guides the weapon is difficult to break. U-Blox is made in Switzerland.
It should be noted that Nvidia and U-Blox have not broken any laws by selling these products. The chips go into the distribution system where they are sold to end users. From there they wind up in Russia or China or Iran.
Washington has tried to crack down on the spread of AI chips in China, but that has effectively meant urging companies not to transfer manufacturing information to China or transfer sensitive AI software.
However, there is little evidence to show that Washington has been able to control the loss of critical components such as the Jetson TX-2. If drastic action is not taken, the Russians, Chinese and Iranians will continue to be able to use advanced AI modules for military and commercial applications.
One reason for this is that the manufacturing of AI products is largely offshored from the United States. This means that high-level cooperation is essential.
It also means that the U.S. is at great risk if AI chip production, especially in Taiwan, is disrupted by war, a blockade or simply a natural disaster. (Taiwan is prone to earthquakes.)
Efforts are now underway to build a new advanced chip foundry in the US, which will be a real help in the future. Even so, it will be years before it actually comes online and Taiwan will continue to prepare for American companies because TSMC and other Taiwanese companies are competitive and highly capable.
For political reasons, the Biden administration is not eager to take advantage of American chip companies. The CHIPS Act, with the U.S. providing massive subsidies, is supposed to help reestablish American manufacturing.
This in itself is a good thing, but it is not related to the spread of AI electronics abroad, especially in China. Unfortunately, the regulatory apparatus in the United States, especially when it comes to DEI (Diversity, Equity and Inclusion), has hindered the rapid use of CHIPS Act funds.
Russia does not have a DEI problem, although it lacks private investment for its chip industry. Indeed, in the big picture, Russia's Achilles heel is its lack of microelectronics manufacturing infrastructure. This happened because Russia was not part of the Western microelectronics revolution.
During the Soviet era, Russia either tried to manufacture its electronics in closed cities like Zelenograd or it offloaded some manufacturing to Eastern Europe, particularly the German Democratic Republic (GDR) aka East Germany.
Like Russia, East Germans and others in the Warsaw Pact were largely isolated from developments in the United States.
In the future, Washington will have to find effective ways to control AI technology or face the consequences. Bradleys and Abrams knocked out by Russia's Lancets in Ukraine represent a real military problem that desperately needs attention.
Stephen Bryan served as Staff Director of the Near East Subcommittee of the US Senate Foreign Relations Committee and Deputy Under Secretary for Defense Policy.
This The article was first published on its Weapons and Tactics substack and is republished with permission.