US Military AI: Project Eagle and a Decade at Risk

Initiatives like Project Eagle, investor activism, and the role of private actors are cited to illustrate the rise of American military AI. The Ukrainian conflict serves as a testing ground, while the period from 2027 to 2037 is presented as the most perilous for the military use of AI.

A 2027-2037 Window Described as the Riskiest

The period from 2027 to 2037 is portrayed as the most dangerous for the military use of AI, with a risk of massive damage to human civilization. Signs are noted of an approaching tipping point for military AI in the United States. It is also estimated that it will soon become impossible to maintain certain systems and protections with human intervention in an increasing number of new aspects of warfare. In this context, the war in Ukraine sees civilian infrastructures targeted, and fears of escalation exist on the European front, including regarding NATO. Meanwhile, AI appears to be gaining strength in areas like cybersecurity, which is considered a military application.

Programs and Actors: From Project Eagle to Anduril

Eric Schmidt is mentioned as leading Project Eagle, which was previously known as White Stork. This operation, described as confidential and launched in late 2023, focuses on the development of suicide drones and low-cost interceptors controlled by artificial intelligence. Anduril, a private company associated with Palantir, is presented as willing to supply weapons to Taiwan. At the same time, the Pentagon is said to be betting on a device named "SI," with no further details provided.

Ukrainian Laboratory and Investment Rush in 2026

The war in Ukraine is seen as a testing ground for the United States, which is analyzing asymmetric warfare there, particularly the use of robotics on the front lines. The simultaneity of the rise of drone swarms, the arrival of advanced military robots, and the outbreak of armed conflicts is viewed as an anticipation of a potential third world war. In 2026, a frenzy of spending by American and European venture capitalists is reported on robotic systems and physical AI, a dynamic illustrated by the activities of figures like Eric Schmidt.

An Expanded Confrontation and Announced AI Armament

AI is presented as likely to increase the probability of confrontations between the United States and China, as well as between the West and the Eastern Bloc or the BRICS. Areas ranging from space to deep-sea mining are described as potentially contested. Over the next ten years, an unimaginable armament of AI is announced on the American side, in a context where global escalation and increasingly AI-mediated technologies would converge in an unregulated and accelerated manner. The author specifies that they are not a national defense analyst. Generative AI took off four and a half years ago, and these remarks are contextualized after the onset of the large-scale invasion of Ukraine in late February 2022. The risks are said to not be limited to the culture of companies like OpenAI, but would also include incentives from public and private powers; venture capital investors are described as having taken partial influence over aspects of technology and AI policy within the American apparatus, up to the Pentagon. The situation is characterized as very serious, with signs of escalation in American preparations to militarize AI.