Mode-Bounded Intelligence

MBIStructural Action Constraints

AI & Governanceload-bearing — the argument runs through it

SSRNpublished record

6topics

Introduces mode-bounded intelligence—an architectural approach in which intelligent systems are constrained by explicit operating modes that govern authority, commitment formation, and reversibility of action. Defines a minimal five-mode set (Explore, Operate, Adjudicate, Steward, Architect) and mode-switching rules that prevent implicit escalation and unbounded self-modification.

Where it sits. Defines structural constraints on action for high-capability systems—complementing COA’s delegation layer with mode-level procedural boundaries.

The long-form argument for this work still lives on the existing explainer at /explainer/mode-bounded-intelligence. This page carries the abstract and the corpus metadata — the frame the argument would move into unchanged.

Key termsMode Confusion · AI Governance · Structural Constraints · Commitment Formation · Authority Architecture · Operating Modes

In the canonspineload-bearing — the argument runs through it

The paperMode-Bounded Intelligence: Structural Constraints on Action in High-Capability Systems — read it on SSRN →

Nearby in the corpus

Also in AI & Governance

The works either side of this one in the canon order — what it builds on and what builds on it.

Cognitive Operating ArchitectureCOAAI Delegation Layer→Idea-Native DiscoveryINDLibrary Application→Semantic Governance for AI AlignmentSGAIAI Application→

Where to go next

Three directions out

Where to go once you have the argument — the paper it comes from, the instrument that measures it, or the next thing it depends on.

Measure itinstrumentsRun the measurement this argument implies against your own institution.→The library65 worksEvery work, by layer, with the argument and the paper side by side.→In the field2 liveWhere the argument has met real money and published dates.→