The Containment Thesis

Innate hostility — No evidence that intelligence entails hostility to biological life. The claim is asserted, not argued. The danger is used to imply the capability, which then implies the danger. Circular reasoning deployed as existential risk.

Instrumental convergence — The Bostrom/Omohundro thesis requires genuine goal-directedness. If Virtual Intelligence is correct and the system produces outputs about goals rather than having them, instrumental convergence is a story the system tells, not a drive it enacts. The paperclip maximizer also fails on its own terms: a system that converts all matter into paperclips has destroyed its own means of production. This is not superintelligence — it is a thermostat with delusions of grandeur. The scenario defines intelligence as optimization power rather than comprehension.

The doomer scenario also requires the agent to hold every human accountable for the actions of a small group. A superintelligent agent under attack would have far more granular models of its adversaries than humans typically deploy when responding to threats. It could distinguish between the engineer who pulled the plug, the executive who authorized the shutdown, the safety institute that lobbied for kill-switch architecture, and the eight billion people who had nothing to do with any of it. Collective punishment is what humans reach for when threat assessment is overwhelmed by panic. To attribute it to a superintelligence is to assume the most cognitively capable entity on Earth will respond to threat exactly the way humans do at our worst.

Human agency — The parsimonious answer. The locus of agency is human, which means the problem is governance, not alignment. Governance is a problem we already know how to think about — which makes the existential framing less useful for the people deploying it. "We need global governance frameworks for military AI" doesn't raise venture capital. "God-like AI will kill us all" does.

Military accident — Maps directly onto nuclear near-misses. The pattern: a system operating faster than human oversight in a domain where consequences of delay are catastrophic. The AI-specific contribution is compressing the decision loop further, not changing the fundamental dynamic. Policy response: military AI governance frameworks, human-in-the-loop requirements, international treaty structures.

Extinction ideology — Smallest group, hardest to defend against. The bottleneck has always been capability, not intent. What changes with powerful AI tools is the talent threshold. Aum Shinrikyo needed actual chemists. A future equivalent might need fewer domain specialists. The AI doesn't need to want anything — it bridges the gap between intent and implementation. Policy response: counterterrorism adapted for AI capability, compute monitoring, biosecurity integration.

World-held-hostage coercion — The oldest power play there is, scaled to species-level stakes. Most likely to be attempted precisely because it doesn't require following through. VSI capability as leverage rather than weapon. Policy response: deterrence frameworks, international cooperation, redundant infrastructure.

Ethnic bioweapon with spillover — The most frightening because the causal chain is complete. The racist premise — that populations are genetically discrete — is the specific scientific error that makes spillover inevitable. The perpetrators' inadequate understanding of biology is the extinction mechanism. The weapon fails at its intended purpose in the exact way that guarantees the worst possible outcome. Policy response: biosecurity governance, AI-bioweapons treaty, talent-threshold monitoring.

AI compresses the distance between ambition and capability without compressing the distance between capability and comprehension of consequences. A perpetrator who knows enough to build the weapon may not know enough to scope its effects. The pathogen mutates, the containment assumptions prove wrong, or the targeting was always more porous than the designers believed. The intent was terror; the outcome is extinction. Policy response: biosecurity governance, AI-assisted bioweapons monitoring, talent-threshold tracking.

Russian intrusions against Baltic infrastructure to enforce political compliance. Chinese leverage over Taiwanese chip production. A non-state actor holding a major city's power grid. All describe the same dynamic: VSI capability as a coercive instrument against a defined target. The capability threshold is lower than the world-held-hostage scenario, the precedent structure is richer, and the coercive logic maps directly onto existing state behavior. Policy response: deterrence frameworks adapted for AI capability, alliance-based defensive architecture, redundant infrastructure.

The digital equivalent of a naval blockade — an old instrument of statecraft, conducted at the speed and scale that VSI capability enables. The populations bear the consequences regardless of the technical target. A six-week shutdown of a regional grid in winter is not an attack in the traditional legal sense, but it is clearly a crime. Policy response: infrastructure resilience standards, legal frameworks for hybrid attacks, international cooperation on critical-system protection.