Open research program

What changes when the organization no longer has to throw context away?

IVA's next research questions sit at the intersection of organizational cognition, accounting architecture, institutional authority, machine memory, evidence, and human accountability.

By Evan FosterPublished
Research agendaAI-era governanceStructural accounting
Core proposition
AI expands what an organization can retain and reason across. Governance determines what that expanded context is allowed to mean and do.

Research lanes

A theory that should produce testable institutional questions.

Context preservation

How does retaining drafts, failed options, informal knowledge, and operating traces change decision quality and organizational learning?

Selection and authority

Which information becomes formal, who controls that selection, and how does the record reproduce Internal Governance Monopoly?

Human-AI governance

Which ledger views, evidence records, and decision rights let people challenge machine-scale reasoning without recreating cognitive bottlenecks?

Structural measurement

How can nonfinancial positions be recognized and compared within a domain without monetization or false cross-domain totals?

Institutional memory

When does comprehensive retention reduce repeated failure, and when do retrieval, authorization, or provenance failures create new liabilities?

Applied instruments

How do Decision Readiness, Value-Based Objectives, and Rule 52 alter implementation, reporting burden, and accountability?

Working propositions

Claims suitable for refinement and testing.

  • The formal organizational record systematically overrepresents domains with established reporting authority.
  • Information discarded for lack of immediate institutional standing retains option value for later decisions.
  • AI lowers the cost of retaining and reactivating organizational context more rapidly than it resolves governance legitimacy.
  • Independent ledger standing improves visibility of cross-domain tradeoffs that consolidated measures conceal.
  • Reporting systems that ignore measurement burden can improve measured targets while reducing total organizational value.
  • Human accountability requires explainable authority and consequence even when machine reasoning exceeds individual comprehension.