Ukrainian and Russian militaries improvise and adapt in a battle of wits (www.cnn.com)

Both Russian and Ukrainian forces are rapidly improvising weapons and tactics, mixing high-tech autonomy with low-tech ingenuity. Ukraine’s “Goronych” group has converted single-seat light planes into pilot-less long-range strike craft—removing the seat for extra fuel and adding batteries to power navigation and comms—for missions up to ~2,000 km that have struck deep inside Russia. Kyiv has also fielded advanced maritime drones capable of >1,500 km range and ~2,000 kg payloads (even mounting multiple-rocket launchers). Russia, meanwhile, is upgrading legacy munitions into guided weapons: glide bombs with ~80 km reach and a new jet-powered guided bomb (UMPB-5) with ~200 km radius have been used alongside massed Shahed-type drone production reportedly exceeding 6,000 units monthly. Both sides also deploy small armed aircraft, dedicated drone units (e.g., Russia’s Rubikon), and hybrid tactics learned in-theater. For the AI/ML community these developments matter because the battlefield is accelerating real-world deployment and iteration of autonomy, sensing, guidance, swarm logistics and contested comms. Technical implications include urgent needs for robust navigation and perception in GPS-denied environments, resilient control under electronic attack, AI-enabled target recognition and routing, and counter-AI measures for detection and jamming. Equally important is the reminder that simple, low-cost adaptations can be strategically significant—so research on lightweight, resilient algorithms and defensive systems will be as critical as high-end autonomy.