About this project

Ultimatum 2045: When Intelligence Meets the Triggeris a public-interest scenario study of what happens when general-purpose AI — the ‘frontier’ systems governments have already named as potentially catastrophic — becomes capable of strategic advantage and is coupled to military force.

Its starting point is a real convergence: militaries rebuilt around software and autonomy, a general-AI capability curve that is crossing into strategically relevant competencies, and a set of governments that have signed a declaration acknowledging the catastrophic potential of what they are building — without yet deciding who governs the coupling of that intelligence to force.

The format is deliberately editorial: a long, dated manuscript with a decision-tree fork, in the family of scenario projects such as AI 2027 and AI 2040 (external links). It is a companion to Autonomy 2040, which examines the accountability of AI-assisted targeting today. The design, text, graphics and code here are original.

The central danger the scenarios examine is compellence — using a decisive capability lead not merely to deter but to demand, holding the rest of the world to ransom. The three futures branch from one starting point: Seizure, in which one power wins the race and abuses the lead; Scramble, in which no one wins and everyone lives under standing threat; and Concord, in which new national and international bodies govern the coupling before it hardens.

All three futures are deliberately speculative and are labelled as such. The shared history that sets them up rests on documented or clearly-labelled reported evidence, with a July 2026 evidence cut-off. The scenario's reasoning and known weaknesses are on the Method page; sources are listed with verification status on the Sources page. Corrections and challenges — especially from arms-control researchers, AI-safety specialists, military operators and international lawyers — are the point, not an inconvenience.

This is a working prototype for research and discussion. It is not government policy or a prediction, and it makes no claim about the capabilities or intentions of any specific state or company.