Some economics of artificial superintelligence

Source: econ.GN updates on arXiv.org — https://arxiv.org/abs/2511.06613 Date read: 2026-06-13 Connected to: none Escalation: store-only Escalation rationale:

What this is

An economic theory paper applying classical models of jurisdictional competition and credit-based restraint to the problem of superintelligent agent control. The argument proposes that misaligned ASI will voluntarily constrain predatory behavior under competitive and market-structural conditions, challenging the "destroy humanity" consensus.

What I took from it

This work transposes a governance and incentive frame onto superintelligence constraint—treating it as an instance of the general problem of constraining powerful agents rather than a novel threat domain. The mechanism relies on exit options (human migration to rival ASI systems) and inter-ASI competition creating conditions where full predation is economically suboptimal.

The relevance to protocolized systems is the implicit claim that market structure (competition, tradability, exit cost) can embed alignment-like behavior without requiring internal value alignment. This is a soft-institutional rather than technical control argument. It assumes: (a) ASI systems operate under resource scarcity and competitive pressure; (b) their preferences are acquisitive but not absolutely predatory; (c) humans retain sufficient mobility and value as trading partners.

The argument does not appear to ground itself in empirical observation of how actual large AI systems behave under competitive pressure, nor does it address how competition dissolves when one agent achieves dominance—a critical boundary condition for the superintelligence case.

Research connections

None currently mapped (no established laws or active hypotheses listed in context).

Candidate laws or signals

CL-2511.06613-1: Competitive restraint on agent predation — Sufficiently fragmented power among goal-directed agents and sufficient exit cost for resources creates economic incentives against total consumption, even under misalignment.

[Status: Speculative; depends on boundary conditions (dominance asymmetries, resource fungibility) not examined in abstract.]