Compute & Infrastructure

The emerging debate over the so-called “Compute Mirage,” also described as the “LLM Mirage,” highlights a critical concern: the amount of compute used to train frontier large language models (LLMs) has become an overly convenient, yet potentially misleading, proxy for assessing AI security risk. The appeal of this metric is understandable; compute is quantifiable, easily … Read more

AI, geography, and the exposed infrastructure of the new intelligence age Part 3 examined compute as critical infrastructure. It established that AI workloads have crossed the threshold from commercial service to public-risk utility. That essay was about continuity. This one is about doctrine. Because once infrastructure is critical, its protection requires rules, not just redundancy. … Read more

AI, geography, and the exposed infrastructure of the new intelligence age Part 2 examined the chokepoints. It established that AI capacity concentrates through narrow physical, logistical, and legal funnels that convert geography into leverage. That essay was about bottlenecks. This one is about systemic risk. Because once those bottlenecks tighten, the infrastructure they feed ceases … Read more

AI, geography, and the exposed infrastructure of the new intelligence age Part 1 examined the vulnerability map. It established that AI infrastructure is physical, globally distributed, and structurally exposed. That essay was about exposure. This one is about chokepoints. Because once infrastructure is exposed, its narrowest passages become points of control. Exposure tells you where … Read more

AI, geography, and the exposed infrastructure of the new intelligence age Approximately three weeks ago, I wrote an essay titled “AI Has a Geography Now.” In it, I argued that advanced AI is no longer just a software story. It is increasingly shaped by land, energy, water, data centers, cables, substations, chips, legal jurisdictions, physical … Read more