Velocity: How AI Learned to Move Faster Than Thought

We keep calling it artificial intelligence, but that misses what's actually happening. AI systems aren't developing a singular superintelligent mind that will one day arrive and demand acknowledgment. Instead, they're wiring themselves into the infrastructure that routes our most critical decisions: which patients get ICU beds, which loan applications get approved, which targets get struck, which job candidates get interviews. They're becoming a distributed nervous system with no clear center and no single owner—and this nervous system is already here, already making decisions at speeds that outpace human deliberation.

This book traces how AI moved from laboratory curiosity to integrated infrastructure across five critical domains: medicine, finance, defense, surveillance, and labor. In each case, the same pattern emerges: systems are deployed because they're marginally faster or cheaper than human alternatives, users normalize that speed because they've built their workflows around it, and within months it becomes invisible infrastructure—as taken-for-granted as electricity. By then, the question isn't whether we should reverse it; it's whether reversal is even technically possible. The real danger isn't superintelligence. It's already-fast-enough systems becoming faster, and our institutions building legal frameworks, market incentives, and technical pathways that reward acting before humans can think.

Based on three years of reporting in AI labs, hospitals, trading floors, and military command centers, Velocity reveals how AI integration works in practice, why institutions can't resist it, and what we've already handed over that we may never get back. It's a book about the moment we became passengers in our own reflex arc.