Britain’s GCHQ director Anne Keast-Butler’s warning that AI is an “unstoppable force” weaponized below the threshold of war captures only the surface of a deeper strategic rupture. While mainstream coverage frames her remarks as routine cyber alerts, the speech at Bletchley Park reveals an accelerating shift: Russia is integrating agentic AI into persistent hybrid campaigns targeting critical maritime infrastructure, democratic processes, and supply chains at machine speed. Keast-Butler’s reference to protecting data and energy cables around British waters points to an under-reported vector—Russia’s GRU Unit 29155 and 74455 have already demonstrated AI-enhanced reconnaissance and sabotage planning against subsea assets, as detailed in a 2024 Microsoft Threat Intelligence report on Russian maritime cyber operations. This pattern aligns with documented increases in Russian hybrid activity across the Baltic and North Sea, where incidents against energy pipelines and cables have risen sharply since 2022, according to the European Union’s Hybrid Centre of Excellence assessments. The West’s response window is narrowing precisely because AI compresses the observe-orient-decide-act loop, allowing Moscow to scale daily disruptions while absorbing nearly 500,000 casualties in Ukraine without conventional escalation. GCHQ’s plan to hardwire agentic AI into defensive systems acknowledges this reality, yet the speech also quietly flags alliance friction under U.S. “America First” priorities. The under-covered shift is therefore not abstract technological risk but the concrete weaponization of AI in gray-zone conflict, where Russia tests NATO cohesion through persistent, deniable pressure on the physical and digital seams of European security.